[FREE PDF sample] Data Mining for Design and Marketing 1st Edition Yukio Ohsawa ebooks

Get ebook downloads in full at ebookname.
com
Data Mining for Design and Marketing 1st Edition

Yukio Ohsawa
https://ebookname.com/product/data-mining-for-design-and-
marketing-1st-edition-yukio-ohsawa/
OR CLICK BUTTON
DOWNLOAD EBOOK
Explore and download more ebook at https://ebookname.com

Instant digital products (PDF, ePub, MOBI) available
Download now and explore formats that suit you...
Visual Data Mining Techniques and Tools for Data

Visualization and Mining 1st Edition Tom Soukup
https://ebookname.com/product/visual-data-mining-techniques-and-
tools-for-data-visualization-and-mining-1st-edition-tom-soukup/
Data Warehousing Design and Advanced Engineering

Applications Methods for Complex Construction Advances
in Data Warehousing and Mining Adwm 1st Edition Ladjel
Bellatreche
https://ebookname.com/product/data-warehousing-design-and-
advanced-engineering-applications-methods-for-complex-
construction-advances-in-data-warehousing-and-mining-adwm-1st-
edition-ladjel-bellatreche/
Investigative Data Mining for Security and Criminal

Detection Jesus Mena
https://ebookname.com/product/investigative-data-mining-for-
security-and-criminal-detection-jesus-mena/
A Higher Bid How to Transform Special Event Fundraising

with Strategic Auctions 1st Edition Kathy Kingston
https://ebookname.com/product/a-higher-bid-how-to-transform-
special-event-fundraising-with-strategic-auctions-1st-edition-
kathy-kingston/
A saga of sea eagles 1st Edition John A. Love
https://ebookname.com/product/a-saga-of-sea-eagles-1st-edition-
john-a-love/
The Java Virtual Machine Specification Java SE 8

Edition Tim Lindholm & Frank Yellin & Gilad Bracha
https://ebookname.com/product/the-java-virtual-machine-
specification-java-se-8-edition-tim-lindholm-frank-yellin-gilad-
bracha/
Intelligent Systems for Automated Learning and

Adaptation Emerging Trends and Applications Premier
Refernece Source 1st Edition Raymond Chiong
https://ebookname.com/product/intelligent-systems-for-automated-
learning-and-adaptation-emerging-trends-and-applications-premier-
refernece-source-1st-edition-raymond-chiong/
Bartimaeus Trilogy Book Three Ptolemy s Gate Jonathan

Stroud
https://ebookname.com/product/bartimaeus-trilogy-book-three-
ptolemy-s-gate-jonathan-stroud/
Netter s Pediatrics 1st Edition Todd Florin
https://ebookname.com/product/netter-s-pediatrics-1st-edition-
todd-florin/
Delta s Key to the Next Generation TOEFL Test Advanced
Skill Practice Book First Edition Nancy Gallagher
https://ebookname.com/product/delta-s-key-to-the-next-generation-
toefl-test-advanced-skill-practice-book-first-edition-nancy-
gallagher/
Data Mining for
Design and Marketing
© 2009 by Taylor & Francis Group, LLC
C7019.indb 1 12/18/08 11:32:57 AM

Chapman & Hall/CRC
Data Mining and Knowledge Discovery Series
SERIES EDITOR
Vipin Kumar
University of Minnesota
Department of Computer Science and Engineering
Minneapolis, Minnesota, U.S.A
AIMS AND SCOPE
This series aims to capture new developments and applications in data mining and knowledge
discovery, while summarizing the computational tools and techniques useful in data analysis. This
series encourages the integration of mathematical, statistical, and computational methods and
techniques through the publication of a broad range of textbooks, reference works, and hand-
books. The inclusion of concrete examples and applications is highly encouraged. The scope of the
series includes, but is not limited to, titles in the areas of data mining and knowledge discovery
methods and applications, modeling, algorithms, theory and foundations, data and knowledge
visualization, data mining systems and tools, and privacy and security issues.
PUBLISHED TITLES
UNDERSTANDING COMPLEX DATASETS: Data Mining with Matrix Decompositions

David Skillicorn
COMPUTATIONAL METHODS OF FEATURE SELECTION
Huan Liu and Hiroshi Motoda
CONSTRAINED CLUSTERING: Advances in Algorithms, Theory, and Applications
Sugato Basu, Ian Davidson, and Kiri L. Wagstaff
KNOWLEDGE DISCOVERY FOR COUNTERTERRORISM AND LAW ENFORCEMENT
David Skillicorn
MULTIMEDIA DATA MINING: A Systematic Introduction to Concepts and Theory
Zhongfei Zhang and Ruofei Zhang
NEXT GENERATION OF DATA MINING
Hillol Kargupta, Jiawei Han, Philip S. Yu, Rajeev Motwani, and Vipin Kumar
DATA MINING FOR DESIGN AND MARKETING
Yukio Ohsawa and Katsutoshi Yada
C7019.indb 2 12/18/08 11:32:57 AM

Chapman & Hall/CRC
Data Mining and Knowledge Discovery Series
Data Mining for

Design and Marketing
Edited by
Yukio Ohsawa
Katsutoshi Yada
C7019.indb 3 12/18/08 11:32:57 AM

Chapman & Hall/CRC
Taylor & Francis Group
6000 Broken Sound Parkway NW, Suite 300
Boca Raton, FL 33487-2742
Chapman & Hall/CRC is an imprint of Taylor & Francis Group, an Informa business
No claim to original U.S. Government works

Printed in the United States of America on acid-free paper
10 9 8 7 6 5 4 3 2 1
International Standard Book Number-13: 978-1-4200-7019-4 (Hardcover)
This book contains information obtained from authentic and highly regarded sources. Reasonable
efforts have been made to publish reliable data and information, but the author and publisher can-
not assume responsibility for the validity of all materials or the consequences of their use. The
authors and publishers have attempted to trace the copyright holders of all material reproduced
in this publication and apologize to copyright holders if permission to publish in this form has not
been obtained. If any copyright material has not been acknowledged please write and let us know so
we may rectify in any future reprint.
Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced,
transmitted, or utilized in any form by any electronic, mechanical, or other means, now known or
hereafter invented, including photocopying, microfilming, and recording, or in any information
storage or retrieval system, without written permission from the publishers.
For permission to photocopy or use material electronically from this work, please access www.copy-
right.com (http://www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC), 222
Rosewood Drive, Danvers, MA 01923, 978-750-8400. CCC is a not-for-profit organization that pro-
vides licenses and registration for a variety of users. For organizations that have been granted a
photocopy license by the CCC, a separate system of payment has been arranged.
Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and
are used only for identification and explanation without intent to infringe.
Library of Congress Cataloging-in-Publication Data
Data mining for design and marketing / Yukio Ohsawa, Katsutoshi Yada.
p. cm. -- (Chapman & Hall/CRC data mining and knowledge
discovery series)
Includes bibliographical references and index.
ISBN 978-1-4200-7019-4 (hardcover : alk. paper)
1. Data mining. 2. System design. 3. Design, Industrial. I. Ohsawa, Y. (Yukio),
1968- II. Yada, Katsutoshi. III. Title. IV. Series.
QA76.9.D343D3828 2008
005.74--dc22 2008044205
Visit the Taylor & Francis Web site at

http://www.taylorandfrancis.com
and the CRC Press Web site at
http://www.crcpress.com
C7019.indb 4 12/18/08 11:32:58 AM

Contents
Chapter 1. Sensing Values in Designing Products and Markets
on Data Mining and Visualizations....................................................1
Yukio Ohsawa
Chapter 2. Reframing the Data-Mining Process................................................ 19

David Bergner and Ozgur Eris
Chapter 3. The Use of Online Market Analysis Systems to Achieve

Competitive Advantage..................................................................... 35
Lihua Zhao, Mark D. Uncles, and Gary Gregory
Chapter 4. Finding Hierarchical Patterns in Large POS Data Using

Historical Trees................................................................................. 57
Takanobu Nakahara and Hiroyuki Morita
Chapter 5. A Method to Search ARX Model Orders and Its Application

to Sales Dynamics Analysis.............................................................. 81
Kenta Fukata, Takashi Washio, Katsutoshi Yada,
and Hiroshi Motoda
Chapter 6. Data Mining for Improved Web Site Design

and Enhanced Marketing.................................................................. 95
Asem Omari
Chapter 7. Discourse Analysis and Creativity Support

for Concept Product Design............................................................ 107
Noriko Imafuji Yasui, Xavier Llorà, and David E. Goldberg
Chapter 8. Data Crystallization with Human Interactions Applied

for Designing New Products........................................................... 119
Kenichi Horie, Yoshiharu Maeno, and Yukio Ohsawa
C7019.indb 5 12/18/08 11:32:58 AM

Chapter 9. Improving and Applying Chance Discovery
for Design Analysis......................................................................... 137
Brett Bojduj
Chapter 10. Mining for Influence Leaders in Global Teamwork Projects......... 149
Renate Fruchter, Shubashri Swaminathan, Naohiro
Matsumura, and Yukio Ohsawa
Chapter 11. Analysis Framework for Knowledge Discovery Related to

Persuasion Process Conversation Logs........................................... 171
Wataru Sunayama and Katsutoshi Yada
Chapter 12. Association Bundle-Based Market Basket Analysis....................... 187

Wenxue Huang, Milorad Krneta, Limin Lin,
and Jianhong Wu
Chapter 13. Formal Concept Analysis with Attribute Priorities........................ 211

Radim Belohlavek and Vilem Vychodil
Chapter 14. Literature Categorization System for Automated Database

Retrieval of Scientific Articles Based
on Dedicated Taxonomy................................................................. 223
Lukáš Pichl, Manabu Suzuki, Masaki Murata,
Daiji Kato, Izumi Murakami, and Akira Sasaki
Chapter 15. A Data-Mining Framework for Designing Personalized

E-Commerce Support Tools........................................................... 235
Timothy Maciag, Dominik Ślęzak, Daryl H. Hepting,
and Robert J. Hilderman
Chapter 16. An Adjacency Matrix Approach for Extracting

User Sentiments.............................................................................. 251
Bin Shi and Kuiyu Chang
Chapter 17. Visualizing RFID Tag Data in a Library for Detecting Latent
Interest of Users.............................................................................. 277
Yukio Ohsawa, Takuma Hosoda, and Takeshi Ui
C7019.indb 6 12/18/08 11:32:58 AM

Appendix A: KeyGraph and Pictorial KeyGraph.................................................. 295
Appendix B: A Maximal Cliques Enumeration Algorithm

for MBA Transaction Data............................................................... 299
C7019.indb 7 12/18/08 11:32:58 AM

Preface
Amidst ever-changing market conditions, the most important thing for companies today
is to build a sustainable process where they can effectively grasp the needs of consumers
in their market, to create products and services that answer these needs, and to appropri-
ately convey information on such products and services to consumers. Because in recent
years consumer needs have diversified and product life cycles have grown shorter, such
a successful process has become increasingly difficult to realize. The ability to imple-
ment this process efficiently forms the resource of sustainable competitive advantage.
In the aspect of engineering, the concept of design has recently been extended to
systems design and is affecting the wide world of business. Systems design is meant
to combine subsystems, which may be basic components for making a product or a
system that may be a complex product, such as a car, or a social system, such as the
market. The mission of a systems designer is to create a new value that has never
been realized by each one of the subsystems by combining the subsystem to form a
new system.
Thus, integrating the aspects of design and marketing, we can declare it is desir-
able to sense and circulate information about the needs of consumers and all stake-
holders as the basic resource for the creation of values. Data mining is an information
technology that is particularly critical to providing this information in two senses.
First, by using data mining, we can deal with vast amounts of data on consumers
and thereby identify valuable patterns. These patterns aid marketers and designers in
detecting the information about needs and circulating the information to coworkers in
comprehensible expressions. Second, by developing visualization tools based on the
computational methods of data mining, designers and marketers can uncover poten-
tial needs from the map of the market even if the needs have never been satisfied, and
apply it to the creation and development of products and services with novel values.
The techniques of data mining are truly growing to redesign business processes and
to develop competitive products and services.
We are aware, unfortunately, that for many businesses data mining has not yet
produced the significant results hoped for. The reason for this does not lie with
only technical issues, such as predictive accuracy or computation speed, or with
the difficulty and the advantage with data mining software used by the companies.
According to our 10 years of interactions with business users of data mining, the
problem is the lack of efficiency in business processes that could produce better
interaction between companies and consumers or among individuals in a company.
If companies can effectively design business processes, data mining can serve as a
powerful engine for their growth.
In various joint research projects with more than 50 companies, we have applied
data mining to marketing and designs of products and services that are among the
most important corporate functions, and we have attained invaluable experience
in business processes utilizing data mining. We gained important knowledge from
C7019.indb 9 12/18/08 11:32:58 AM

this experience: we noticed a positive “spiraling process” in which people inside an
organization learn something new from data, devise business scenarios based on
this knowledge, and learn even more new things as these scenarios create valuable
products if put into practice. The topics presented in this book are not only about
traditional evaluation criteria of tools for data mining such as predictive accuracy or
computation speed; we also offer suggestions from the perspective of creativity, and
a surprise factor creates valuable products when devising new business scenarios. It
is our sincere hope to be able to add something new to the research and business of
this book’s readers.
Yukio Ohsawa and Katsutoshi Yada

On the bridge between academia and business
C7019.indb 10 12/18/08 11:32:59 AM

Editors
Yukio Ohsawa is an associate professor in the Department of Systems Innovation,
School of Engineering, The University of Tokyo. He earned his Ph.D. in com-
munication and information engineering from The University of Tokyo. He also
has worked for the School of Engineering Science at Osaka University (research
associate, 1995–1999), Graduate School of Business Sciences at University of
Tsukuba (associate professor, 1999–2005), and Japan Science and Technology
Corporation (JST researcher, 2000–2003). He initiated the research area of
chance discovery—defined as “discovery of events significant for decision mak-
ing” in 1999—and a series of international meetings (conference sessions and
workshops), such as the fall symposium of the American Association of Artificial
Intelligence (2001). He edited the first books on Chance Discovery (2003) and
Chance Discoveries in Real World Decision Making (2003, Springer–Verlag) and
special issues in international and Japanese (domestic) journals. Chance discovery
is growing: articles have been published in international journals—Journal of
Contingencies and Crisis Management (2001), New Generation Computing (2003),
New Mathematics and Natural Computing (2005), and Journal on Soft Computing in
conjunction with the special issue on Web intelligence (2006). He is on the editorial
board of the Japanese Society of AI and the planning board of New Generation
Computing, and he is the Technical Committee chair of the IEEE-SMC technical
committee of Information Systems for Design and Marketing.
Katsutoshi Yada is a professor of management information systems in the Faculty

of Commerce, Kansai University, Osaka, Japan. He was previously an assistant
professor in the Department of Business Administration, Osaka Industrial University,
1997–2000, and Faculty of Commerce, Kansai University, Osaka, Japan, 2000–
2007. He was a visiting scholar at the marketing division of the Graduate School of
Business, Columbia University, New York, 2006–2007. He received his M.A. and
Ph.D. in business administration from Kobe University of Commerce, Hyogo, Japan,
in 1994 and 2002, respectively. He received the Encouragement Award of Japanese
Society for Artificial Intelligence (2003) and Session Best Presentation Award at
SCIS and ISIS (2006). His present research interests include data mining for market-
ing and information strategy concerning data mining. He has been chairman and
program committee member in many international data mining conferences. He is a
member of IEEE, AMA, and AMS.
C7019.indb 11 12/18/08 11:32:59 AM

Contributors
Radim Belohlavek, Ph.D. David E. Goldberg, Ph.D.
State University of New York Department of Industrial and
at Binghamton Enterprise Systems Engineering
Binghamton, New York, and University of Illinois at
Palacky University Urbana-Champaign
Olomouc, Czech Republic Champaign, Illinois
David Bergner, Ph.D. Gary Gregory, B.Sc., M.B.A., Ph.D.

Division of Space Biosciences School of Marketing
NASA Ames Research Center University of New South Wales
Moffett Field, California New South Wales, Australia
Daryl H. Hepting, Ph.D.

Brett Bojduj Department of Computer Science
Department of Computer Science University of Regina
California Polytechnic State Regina, Saskatchewan, Canada
University
San Luis Obispo, California Robert J. Hilderman, Ph.D.
Department of Computer Science
Kuiyu Chang, Ph.D. University of Regina
Mosma Research Regina, Saskatchewan, Canada
Mosma Corporation
Singapore Kenichi Horie, M.S.
School of Engineering
The University of Tokyo, Japan
Ozgur Eris, Ph.D.
Franklin W. Olin College Takuma Hosoda
of Engineering Department of Systems Innovation
Needham, Massachusetts Faculty of Engineering
The University of Tokyo
Renate Fruchter Tokyo, Japan
Civil and Environmental Engineering
Department Wenxue Huang, Ph.D.
Stanford University Generation-5 Mathematical Technologies
Palo Alto, California Toronto, Ontario, Canada
Kenta Fukata Daiji Kato

Graduate School of Engineering Coordination Research Center
Osaka University National Institute for Fusion Science
Osaka, Japan Gifu, Japan
C7019.indb 13 12/18/08 11:32:59 AM

Milorad Krneta, Ph.D. Masaki Murata
Generation-5 Mathematical National Institute of Information and
Technologies Communications Technology
Toronto, Ontario, Canada Kyoto, Japan
Limin Lin, Ph.D. Takanobu Nakahara, M.A.

Generation-5 Mathematical School of Economics
Technologies Osaka Prefecture University
Toronto, Ontario, Canada Osaka, Japan
Xavier Llorà, Ph.D. Asem Omari, Ph.D.
National Center for Supercomputing Databases and Information Systems
Applications Institute of Computer Science
University of Illinois at Urbana- Heinrich-Heine University
Champaign Duesseldorf, Germany
Champaign, Illinois
Lukáš Pichl, Ph.D.
Timothy Maciag
Division of Finite Systems
Department of Computer Science
Max Planck Institute for the Physics
University of Regina
of Complex Systems
Regina, Saskatchewan, Canada
Dresden, Germany
Yoshiharu Maeno, M.S., Ph.D.
Akira Sasaki
School of Business Sciences
Advanced Photon Research Center
The University of Tsukuba
Japan Atomic Energy Research
Ibaraki, Japan
Institute
Naohiro Matsumura, Ph.D. Kyoto, Japan
Graduate School of Economics
Osaka University Bin Shi, B.Eng., M.Eng.
Osaka, Japan School of Computer Engineering
Nanyang Technological University
Hiroyuki Morita, M.A., Ph.D. Singapore
School of Economics
Osaka Prefecture University Dominik Ślęzak, Ph.D.
Osaka, Japan Infobright
Toronto, Ontario, Canada
Hiroshi Motoda, Ph.D.
Asian Office of Aerospace Research Wataru Sunayama, Ph.D.
and Development Hiroshima City University
Air Force Office of Scientific Research Hiroshima, Japan
U.S. Air Force Research Laboratory
Tokyo, Japan Manabu Suzuki
Department of Computer and
Izumi Murakami Information Science
Coordination Research Center Graduate School of Arkansas Tech
National Institute for Fusion Science University
Gifu, Japan Russellville, Arkansas
C7019.indb 14 12/18/08 11:32:59 AM

Shubashri Swaminathan Takashi Washio, Ph.D.
Civil and Environmental Engineering The Institute of Scientific and
Department Industrial Research
Stanford University Osaka University
Palo Alto, California Osaka, Japan
Jianhong Wu, Ph.D.

Takeshi Ui, M.S. Laboratory for Industrial and Applied
Toppan Forms Mathematics
Tokyo, Japan Department of Mathematics and
Statistics
York University
Toronto, Ontario, Canada
Mark D. Uncles, B.Sc., Ph.D.
School of Marketing Noriko Imafuji Yasui, M.S., Ph.D.
University of New South Wales Department of Industrial and
New South Wales, Australia Enterprise Systems Engineering
University of Illinois at
Urbana-Champaign
Vilem Vychodil, M.Sc., Ph.D. Champaign, Illinois
State University of New York
at Binghamton Lihua Zhao, M.Sc., Ph.D.
Binghamton, New York, and School of Marketing
Palacky University University of New South Wales
Olomouc, Czech Republic New South Wales, Australia
C7019.indb 15 12/18/08 11:32:59 AM

1 Sensing Values in
Designing Products and
Markets on Data Mining
and Visualizations
Yukio Ohsawa
Contents
1.1 Introduction: Design and Marketing as the Process to Value Sensing..............1

1.2 Two Cases: Marketing and Design with Communications
on Visualized Data.............................................................................................3
1.3 Data Collection as the Root of Everything: Evidence
of Objectivity and Subjectivity..........................................................................8
1.4 Scenario Maps as a Basis of Chance Discovery................................................9
1.5 Meta-Cognition as the Effect of Data Visualization........................................ 10
1.6 Gap Between Advertisers and Designers: Results
of Visualizing Messages.................................................................................. 11
1.6.1 The Results and Implications................................................................ 12
1.7 Conclusions...................................................................................................... 15
References................................................................................................................. 16
1.1 Introduction: Design and Marketing

as the Process to Value Sensing
The application of data mining to business has grown explosively in the last decade.
For example, the following provide benefits to real companies:
Predicting sales quantity of commercial items1,2

Finding the associations among commercial items, such as “if a customer buys
bread, he or she tends to buy butter at the same time”3
Predicting stock prices4
Predicting network faults5
These analyses have been regarded as methods for showing objective conclusions
because they are based on real data: facts according to a well-arranged sensing sys-
tem such as error detectors in computer/telephone networks, news of stock prices,
C7019.indb 1 12/18/08 11:32:59 AM

2 Data Mining for Design and Marketing
points of sale, and registers in a supermarket, for example. As a result, the decision
makers in business tend to rely on the “trustworthy” results obtained by tools of data
mining and make decisions such as “we can recommend butter if a customer orders
bread” because of patterns demonstrated in the past.
A problem with this approach occurs because humans often outperform the algo-
rithm of data mining. For example, when a student orders a book by entering the
title in the search engine of an online bookshop site, he may find a list of five books
recommended. However, the student already possesses two of the books and he is
already planning to purchase the other three. As a result, the system cannot give him
any new information, because he has been keenly sensitive to newly published books
and knows more than the search results.
Some recommendation systems have introduced devices to suggest items that are
not far from but not too close to the request based on the product order history.6 For
example, if a customer orders bread, recommending wine may be more reasonable
than recommending butter, because the customer is used to buying the set {butter,
bread} for breakfast and considers bread as food as a different category, such as for a
side food to wine. A strategy for recommending this may be to show item X, which
does not co-occur with “bread” by high conditional probability (or confidence of the
association rule of “bread → X”) as 5%, but does by 1% or 2%. Some online book
shops have already introduced heuristics to present not-close not-far items.
When discussing not-close not-far items, consider that there are a huge number of
proverbial trees within 10 meters of the mountaintop in comparison to those within
1 meter. We need additional strategies for bringing customers’ attention to a mean-
ingful subset of items obtained by the not-close not-far items heuristics, because
the “not-close” condition involves a huge number of candidate items. Otherwise, the
recommendation may overwhelm customers with too much information. Marketers
instead want a strategy for computing the utility of each item for the customer. There
is a question we have to answer first, however: “What is the utility of an item?” To
answer this question, consider how the customer is planning to use the item. For
example, the high utility of “wine” with bread can be understood by considering that
the customer takes bread as a starting food for dinner and the wine to serve with a
steak dinner. Thus if a customer buys bread with wine, it means he or she may also buy
beef to grill and vegetables for side dishes. On the other hand, to evaluate the utility of
“butter” with bread, we should consider the scenario of eating bread for breakfast
and having butter with the bread and assume the customer will not buy additional
expensive items. Thus a marketer has to notice a scenario to connect a commercial
item to a situation (such as dinner or breakfast) in evaluating the item’s utility.
This principle relates also to finding the good design of a product. In a develop-
ment section of our Japanese sponsor, equipment to find scratches on CCD film
(Figure 1.1, left side) has been designed. A film sample is placed on the right-hand
edge of the belt and conveyed on the left. On the way, lights are shone onto the sur-
face of the film. At the first point where the film meets the light (I), cameras take
an image of the light’s reflection. Then, running some more distance, the film meets
the second light (II), at which time the image of the light coming through the film
is taken by the cameras (II). Both of the images taken by the two camera sets are
sent to the computer (image processor) and the scratches on the film are diagnosed.
C7019.indb 2 12/18/08 11:33:00 AM

Sensing Values in Designing Products 3
Line sensing
Image processor
cameras (II)
Remote
controlling
station
Transparency Light (II) Line sensing
checker cameras (I)
Light (I)
Reflection
checker
Figure 1.1 A problem in product design: how can we mix and combine multidisciplinary
knowledge to design and redesign these products?
To redesign this system, it would be necessary to choose the most essential compo-
nent or component set that would provide an effective method for such an improve-
ment. For example, if there are new types of scratches that are hard to distinguish
from each other, a new combination of existing components, or, in a complex case,
new components altogether, might be required. Considering the end result is impor-
tant for understanding the merit of each new design.
Similar requirements for user scenarios are important to such products as clothes
(Figure 1.1, right side). Before selling a product, the marketer needs to predict the
consumer’s response (the extent of satisfaction) by drawing a scenario such as “the
consumer will wear this clothing and go out for lunch on a cold day in autumn.” This
scenario is useful for designing clothing as well as marketing the clothing. If a textile
is found to be suitable for making clothing to wear for going out to lunch on a cold
day, and if there is consumer demand to go out for lunch in winter, the clothing would
be accepted by consumers if designed well and the textile will be bought by apparel
industries where the cloth is designed and sold.
These scenarios outline the relation of a person who is a customer or a decision
maker in business, a group of coworking staff members of a company, or the huge
number of consumers to events and situations in the environment. Value sensing,
to feel associated with things in the environment, has been defined as a particu-
lar dimension of human awareness in the literature of educational psychology.7 It is
meaningful to extend this concept to an aspect of creativity in business. The “value”
here can be dealt with as a new relationship to the social and natural environment,
which business workers and customers create from their interaction via their prod-
ucts and services and through which they eventually redesign the market.
1.2 Two Cases: Marketing and Design with

Communications on Visualized Data
Let us continue with the example of textile. A textile company has been strongly
seeking to develop and sell new products (i.e., new kinds of textiles to be accepted by
the market). The company is well aware that the three major parts of its market are
C7019.indb 3 12/18/08 11:33:00 AM

Submarket 1: Submarket 3:
Textiles for “worn-out” “Dense textile for
style clothes, for outdoor neat clothes, e.g.,
use clothes for blouse”
Submarket 2:
Already popular
for suits
Figure 1.2 How can we combine multidisciplinary knowledge for satisfying the consumers/
users?
related to textiles for business suits (Submarket 1), business underwear (Submarket 2),
and casual wear (Submarket 3). Typical designs of these clothes are shown in
Figure 1.2.
Although the company has mature markets of these kinds of products, it wanted
to develop new markets starting from a niche product (i.e., a product that may not
be popular for the time being but can expand the market in the future). The com-
pany started from data collected in exhibition events in which the product (textile)
samples were arranged on tables, and customers representing apparel companies
picked samples and left a list of preferred samples. Previously, such a list had been
used only as an order card on which the textile company could send samples to the
customers. However, now the company has found that the same list, if put into an
electronic dataset, could be a source of ideas for designing and marketing new prod-
ucts. In comparison with data on past sales, the data from exhibition events include
customers’ preferences for new products that had not yet been sold.
The marketers of this textile company visualized the data using decision trees,
correspondence analysis, etc., and looked for the best tool for achieving this goal.
They decided on KeyGraph,8–10 which shows (1) clusters of frequent items in the data
(i.e., popular item sets ordered by the same customer) and (2) items bridging clusters
(i.e., that appear rarely but appear in the same baskets as items in multiple clusters).
(The reader is referred to Appendix A for more information about KeyGraph and
Pictorial KeyGraph, a novel version of KeyGraph with an interface to attach pictorial
information onto the scenario map obtained by KeyGraph.8)
The scenario map was obtained as in Figure 1.3, where the black nodes linked by
black lines show the clusters of (1) mentioned previously, and the red nodes and the
C7019.indb 4 12/18/08 11:33:01 AM

Figure 1.3 Scenario map from KeyGraph for the data on textile exhibition party.
red lines show items of bridges in (2) and their co-occurrence with items in the clus-
ters, respectively. This figure, however, is difficult for the marketers to understand
because of the complex names of products (i.e., complicated sequences of numbers
and letters corresponding to the features of the textile materials). The marketers
therefore attached the product samples, as in Figure 1.4, to sense the smoothness,
thickness, and colors themselves.
“Dense textile
for neat clothes, e.g.,
clothes for blouse”
Textiles for “worn-out”

style clothes, for out- The chance? Yes! The top
door use 13 of 800 products.
New types of corduroy
Already popular
for suits
Figure 1.4 Marketing as a problem of value sensing on the scenario map.
C7019.indb 5 12/18/08 11:33:03 AM

The meeting by the 10 members of the marketing section ran as follows:
1. First, the marketers noticed the clusters corresponding to popular item sets
mentioned in (1). Three marketers who were experts in women’s blouses
noticed the meaningful cluster at the top of Figure 1.4 (i.e., “dense textile
for neat clothes [e.g., clothes for blouse]”); three others noticed the clus-
ter at the right of Figure 1.4 corresponding to business suits. Two others
noticed the popular item, not linked to any clusters of (1) via black lines,
in the left of the diagram corresponding to materials of casual clothes.
These correspond to the submarkets in Figure 1.2.
2. Second, a marketer who had been working for the company more than
15 years paid attention to the relationship of the other marketers’ opin-
ions with the three submarkets above. He pointed out that women using
daily business wear (a combination of suits and blouse) do not like to
wear the same kind of clothing for the weekend but, instead, desire to
change into casual wear made of such a textile as in the left-hand cluster
of Figure 1.4.
3. Based on the scenario found in Step 2 above, the marketers paid attention
to items in the large meta-cluster, which is the combination of the clusters
at the top, on the right, and the popular items on the left. These “between”
nodes are red nodes (i.e., items lying between popular clusters), on which
the marketers finally calculated a scenario to design a new semicasual
(or semiformal) cloth that women can wear both to the workplace and to
lunch or dinner with friends. As a result, the material indicating the red
node, in the dotted circle in Figure 1.4, produced a historic sales hit of the
13th highest sales among their 800 products (previously, the sales of new
products had been normally lower than the 100th highest).
This example (see previous work for more details on the technical aspect of this
case11) implies three important principles. First, data collected on the users’ (in this
case, marketers’) sense of value (i.e., the relationship of target events to their interest)
leads the company to success in marketing by designing products to sell. Second,
the visualization tool presents a workplace in which coworkers bring their multidis-
ciplinary expertise to bear and create new business scenarios from the teamwork.
Third, not only visualization of the raw data, but also devising a new interface for
showing it to suitable users, as done with attaching real textile pieces, plays a role as
a strong support of creative decision in business.
Here is another example. The company’s aim is to redesign a machine (Figure 1.1,
left side). As mentioned earlier, this machine is a complex system composed of cam-
eras, light, computer, software, and a mechanical belt conveyer. The company’s
development section employs technicians from various engineering domains such as
optical engineering, computer science, and mechanical engineering. It is difficult for
the manager of this section to organize projects to redesign this complex machine,
because it is essential to combine the expertise of these professionals for making
the whole system run collaboratively. In other words, looking over the variety of
C7019.indb 6 12/18/08 11:33:03 AM

Speed of image
Image sampling
processing
software
Control of
belt conveyer
Lens controller
“spottiness” “bubbles” and resolutions
“line-like”
Slits Light conditions
The CCD
problems
Figure 1.5 The visualization of business reports on the technicians’ communications with
users.
knowledge from a bird’s-eye viewpoint is required, but is hard for a manager to do

alone.
To cope with this problem, the manager collected the customer reports written
by the technicians in this section. These reports have been based on the technicians’
communications with the users of the machine, such as film manufacturers (because
the users are also technicians who are highly trained and have technical knowledge,
this company did not send the sales force team but the technicians to the custom-
ers’ working place). Because the collected reports included technical terms hard to
understand for the manager, who is the decision maker who decides what products to
develop in which way to redesign the machine, the reports hardly reflected the deci-
sion consensus of the section.
However, after viewing the collected reports and depicting each frequent word
by a black node, the co-occurrence of these frequent words in the same reports by
black lines—cluster of (1) mentioned in this section—and rare words co-occurring
with words in multiple clusters—items of (2)—the overall structure of reports came
out clearly, as in Figure 1.5. The major parts of this diagram are as follows. (A) The
upper-left part of the figure corresponds to software techniques for image process-
ing, (B) the upper-right and center of the figure are about mechanics for moving the
belt conveyer and the lens of cameras, and (C) the bottom shows the variety of light
conditions. The words about scratches are located in the bottom left, between the
image processing and the light conditions. The manager organized meetings with
the members of his section to share and look at the figure as a scenario map. The
participants were allowed to draw auxiliary lines, to explain their understanding of
this figure’s structure and to put annotations with letters expressing scenarios they
thought of, on the printed-out scenario map.
C7019.indb 7 12/18/08 11:33:04 AM

After discussions, they understood that different types of scratches can be detected
by using different types of techniques. For example, the scratch called “spottiness”
should be distinguished from others by combining special light conditions and image
processing software. As a result, this team registered for new patents and now is the
top sales brand in the market of film scratch inspection. The reader is referred to
details of this case’s analysis elsewhere12 and its extension to creating new ideas from
patent documents in previous work.13
1.3 Data Collection as the Root of Everything:

Evidence of Objectivity and Subjectivity
We can summarize the cases above as Figure 1.6. The process started from collec-
tion of data, which is based on keen attention to the emergence of new values. For
example, in the case of the textile company, it started from the motivation to develop
new textile products and then collect preference data from product exhibition events.
In the case of the film-scratch inspecting machine, the manager had a severe problem
in developing and redesigning products from his communication with technicians or
from the technicians’ communications with customers (the machine users). Then,
they introduced the visualization of data using suitable tools, which happened to
be KeyGraph. Looking at the visualized diagram, the map of their markets, they
discussed the scenarios of developing and selling products. Through this discussion,
the marketers in the textile company and the technicians in the machine company
obtained novel scenarios that led them to success.
Attention to
Sense new values novel values
and new events
Interaction with Next cycle
environment Visualize thoughts,
words, eye tracks,
etc for meta-
cognition
Visualize scenario
Actions maps and rules
che from collected
ese
for data
20$
“20$ cheese is not a FSI,

but is important!”
“che
e
“che se, bre
e se, ad
chip -> bu
s -> tter”
win
e” Attention to
hidden values
Figure 1.6 Spiral process with white board: Data as objective evidences are the results of
subjective interests, and vice versa. Both data are visualized and always shown to user(s).
C7019.indb 8 12/18/08 11:33:05 AM

A point not mentioned previously is that the discussants were also visualizing
their thoughts during the meetings. In the textile case, they were writing the dis-
cussion outlines on a whiteboard. In the machine company, words were typed into
KeyGraph during communications and were visualized on their co-occurrences in
the same sentences as in the center of Figure 1.6. Objective data from the real world
or (not exclusive) subjective data from their thoughts were visualized. Thus, partici-
pants of group decisions were always aware of their own priorities and interests and
understood which of their targets were linked to the teammates’ main interests and
how these links could produce benefits. We have called the process a “double helix”
because humans and computer always run up the spiral, where humans give data to a
computer and a computer gives visualized results to humans.14 However, in this chap-
ter, we may call this the whiteboard process, because successful users really referred
to two or more scenario maps obtained throughout the collaboration process.
1.4 Scenario Maps as a Basis of Chance Discovery

The common point between the two cases is that they developed new attention to a
small part of their market, which had not been previously considered. In the case of
the textile company, the corduroy material that finally led to a sales hit had rarely
been ordered by customers. In the case of film-scratch inspection machines, techni-
cians had not noticed that some scratches were distinguished by the combinations of
light and image processing software. And, on the visualized maps, they found these
items to be a significant part of a new scenario.
To understand this effect, refer to Figure 1.7, which visualizes a customer’s behav-
ior in a supermarket, as a scenario map. Suppose the customer is used to buying only
liquor and the store manager finds it difficult to recommend he buy food in this store.
To attract such a poorly motivated customer, the store manager placed an expensive
($20) cheese on the shelf close to the liquor. The customer became interested in
Figure 1.7 A scenario map, showing frequent episodes and an between-lier product.
C7019.indb 9 12/18/08 11:33:06 AM

the nice-looking cheese. However, he did not buy it because of the high price. The
manager was disappointed and thought of getting rid of the cheese from the shelf.
However, this customer walked up to the shelf of snacks and chose to buy several
low-cost items such as chips and cheese-type snacks. This action was the result of
the manager “suggesting” the taste of cheese, which then motivated the customer’s
interest in things to eat with beer and wine.
By thinking of such scenarios and looking at a scenario map, we can interpret
the behavior of such a customer and understand the role of the $20 cheese, because
the cheese affected the customer’s decision to buy other items than just the liquor
he was accustomed to buying. This is good thinking on the part of the supermarket
manager, who needs to decide which items to arrange on the shelves. The $20 cheese
seldom appears in the point of sale data, because visitors tend to just pick it out and
return it to the shelf, as this customer did. Items and events that significantly influ-
ence people’s decisions may be rare in the data, insomuch as we collect data on the
final decision of consumers/customers/product users only.
KeyGraph, as in the examples shown previously, assisted business decisions to
take advantage of rare or novel items, because the information presented by the
graph was a map of the market, as simplified in Figure 1.7. Thus the diagram had
to have clusters of (1) in Section 1.2, corresponding to the liquors and the snacks
in Figure 1.7, and items as bridges in (2) of Section 1.2 between multiple clusters,
corresponding to the $20 cheese. For details of KeyGraph algorithms, the reader
is referred to previous work;8 this tool was invented 10 years ago and is prevalent
in industries mainly in Japan. We started studies on chance discovery under the
definition of “chance” as an event significant for people’s decisions from inter-
national sessions in 2000 (i.e., IECON2000, KES2000, and AAAI fall symposia
2001) and journal special issues (Journal of Crisis Management Contingencies
in 2001, New Generation Computing in 2002 and 2003, New Mathematics and
Natural Computing in 2006, and Information Sciences in 2008). For all articles
and books, we stand on the simple principle that a decision means choosing one
from multiple future scenarios of actions and events. Based on this principle,
a chance can be regarded as an event at the cross-point of scenarios, because
people are presented with a new stimulus to choose one from multiple scenarios.
The bridging events or items that KeyGraph presented have been candidates of
chances. This way of looking at diagrams, as in Figure 1.6, was more useful than
the diagrams themselves, according to the real users of KeyGraph working on
marketing and design.
1.5 Meta-Cognition as the Effect of Data Visualization

Presented here is an experiment in which we hired a marketer who had been work-
ing on textile materials design. We had him wear an eye movement tracker made
of glasses with two cameras: one detecting eye movement and the other tracing the
target object, so we could acquire his eye movement on the KeyGraph diagram.
First, he looked at the two major clusters of popular items, as in Figure 1.8. Then he
glanced for 1 second at the item pointed to by the arrow in Figure 1.8; for 10 seconds
after that, his eyes traced the dotted line. During the 10 seconds, the eyes moved
C7019.indb 10 12/18/08 11:33:06 AM

Marketing scenario for knit, for summer
Popular 1
Popular 2
Figure 1.8 Eyes catch scenarios from a data-based scenario map.
quickly like a saccade and then moved slowly. After this specific movement, his eyes
drifted all over the diagram for 5 minutes.
I asked him if he found any “chance,” i.e., event significant for his decision, in
this experiment. He said he found none. Then I showed the dotted lines in Figure 1.8
and told him these were extracted from his own cognitive process. He was deeply
impressed and said, “Yes, I looked at the interesting textile pointed [to] by the arrow.
This was a material for making thin clothes such as women’s business-appropriate
underwear. However, in this experiment, I noticed that the color and the smooth-
ness in touching this textile material may be rather suitable for casual wear. I shall
propose a new way to sell this material.” As a result, this item achieved a high sales
performance as a material for casual wear.
This example clarifies that the user’s attention to a “chance” can be further stimu-
lated by meta-cognition (i.e., cognition of one’s own cognition).15 This encourages
users to talk about scenarios coming to their minds, through which the words may
stimulate meta-cognition and accelerate the spiral process of chance discovery.
1.6 Gap Between Advertisers and Designers:

Results of Visualizing Messages
Throughout this section, we focus attention on people who make a living directly
or indirectly from the design of products. They are the products’ (1) designers, (2)
advertisers, and (3) evaluators from the viewpoint of working outside the manufac-
turing company (i.e., users and professional evaluators). We expect people in (1) to
work on strong inspiration and motivation to create new products, which may not be
a motivation shared with people in group (3). A hypothesis is that people in (2) may
C7019.indb 11 12/18/08 11:33:07 AM

be a good bridge between people in (1) and (3). If this hypothesis is correct, we can say
that the concept (the reason for giving priority to a product’s unique features) behind a
design may not be well understood by people outside the manufacturing company.
We do not mention “users” for item (3) because our goal is to see how the design-
ers’ intentions trickle down from in to out of the community of people engaged in
manufacturing. However, we cannot provide statistic validation of our hypotheses
because we cannot set a hypothesis in a simple form such as “product X is designed
under the concept of Y, but the user does not say they relate X and Y” because of the
vocabulary gap16 between designers and users in calling. In this sense, interviewing
different categories of people about each product is not meaningful.
However, we may expect designers to say “products X1 and X2 are designed
under the concept Y,” whereas the user says “X1 is felt to be Y1, and X2 is also like
Y1,” where Y is not equal to Y1. They are talking in different vocabularies, but the
two real worlds of products expressed can be regarded as similar, considering the
similarity in the co-occurrence of those products. On the other hand, if the user says
“X1 is felt to be Y1, but X2 is like Y2,” where Y1 and Y2 are different, his or her
world is different from the world in the designers’ minds.
Instead of statistic validation, we chose to compare the positions of products of
people in different categories. We collected the data on cellular phones that won the
Good Design Award between 2003 and 2005,17 and we analyzed the data as follows:
1. For each year between 2003 and 2005, we collected the three kinds of
data for designed products (cellular phones):
a. The documented designers’ comments
b. The documents from the Web pages where the product is advertised
c. The referees’ answers to questionnaires about each product’s design
2. For the data set of each year, we visualized the positions of products in the
map visualized with KeyGraph.
3. We measured to what degree each company’s products are isolated in the
map from those of other companies.
1.6.1 The Results and Implications

The results of Pictorial KeyGraph (see Appendix A) are shown in Figure 1.9,
Figure 1.10, and Figure 1.11, respectively, for the three data sets below, for the award-
winning cellular phones in 2004.
Data a took the form as:
Da = Product 1: I tried to smooth and …

Product 1: feature 1) Smooth and soft …
Product 1: feature 2) Easy use of …
Product 2: This design is dedicated to …
Product 2: feature 1) The camera is good…
Product 2: feature 2) Photos can be sent…
… (Equation 1.1)
C7019.indb 12 12/18/08 11:33:07 AM

Company A
Classy
Stylish
Company B
One hand
Touch
feeling
Handy
Visibility
Company C
Eﬃciency
Company D
Compact
Figure 1.9 Pictorial KeyGraph applied to the comments of designers of the best designed
cellular phones in 2004. The isolation degree is 11.0 for the four major companies.
Figure 1.10 Pictorial KeyGraph applied to the advertisements on the Web for the best
designed cellulars in Figure 1.9. No companies exist of which the products form isolated
clusters. The isolation degree is 4.0 for the four major companies. Do not read Japanese char-
acters: the isolation degrees are the point of this figure.
C7019.indb 13 12/18/08 11:33:09 AM

Company C
Company A
Company D
Company B
Figure 1.11 Pictorial Key Graph applied to the judgments of referees for the best designed
cellular phones. No companies exist of which the products form isolated clusters. The isola-
tion degree is 7.49 for the four major companies.
Then data b were taken from the Web pages of companies’ cellular advertise-
ments, and the formats were similar to Da. Data c were
Dc = Beautiful: 1, 3, 15, 17, 19…

Novel: 2, 5, 15, 21, …
Useful: 1, 2, 15, … (Equation 1.2)
where the numbers show the identification numbers of products and the lines are the
item evaluation criteria in the questionnaire. In Equation 1.2, product numbers 1, 2,
15, etc., were useful to the referees. Each year, we had 30–58 well-designed cellular
phones, and each product was designed by between one and five designers. The referees
judged mostly just by investigating the products, rather than hearing from designers.
These data can be visualized by KeyGraph to show the positions of products in the
map of cellular phones. As marked by the shadows we show in Figure 1.9, the designers’
comments are clearly separated in areas of products designed by different companies.
In Figure 1.9, each company’s area is featured by words representing the design-
er’s motivations, which are rare in the comments but connect the products of the
company. These are ambiguous words, but can be understood by looking at the
embedded pictures. For example, we find “compact” and “efficiency” for company
D. This combination of words is hard to understand, but the unique look of the small
products in the pictures shows rich functions such as movable and television-like dis-
plays in a small surface of a cellular. We can understand these products are compact
and efficient. As well, we find “classy” as a part of the design concept in company B.
This is another word difficult to understand, but the silver-colored body of the cellular
phones in the area for company B in Figure 1.9 expresses the word’s meaning.
This result shows that the designers in a company share some concepts, but that
their low frequency may hide their existence. That is, if the concepts are expressed
C7019.indb 14 12/18/08 11:33:10 AM

as frequently in words as in the figure’s islands of black nodes, the company’s staff
and consumers would have had many opportunities to express those words. Because
of the rareness, we can really expect those concepts to be discounted by people living
inside and outside the company. This expectation is clearly supported by introducing
a new criteria factor of how clustered the product set of each company is. We call
these criteria isolation degree, defined by the procedure below for each map:
1. Take N companies whose numbers of products on the map are the largest.
Call them major companies.
2. For each major company j (1 < j < N), surround as many products from
company j as possible with one closed rim without crossing with itself or
other rims. If any products of company j are not surrounded in the rim,
make a new rim until all products of company j are surrounded by some
rim. The set of products in a rim is called a rim-cluster.
3. Compute Iso, the isolation degree of the map with the following equation:
Iso = The number of products produced by the N majors

/Σ j=1 to N {the number of clusters of j-th major company}
(Equation 1.3)
Intuitively, Iso is large, if products from each company make a cluster that is iso-
lated from other companies’ clusters.
According to Figure 1.9, we find that the designer’s comments in the same com-
pany tend to form a group, sharing some concept for design (shown as the letters in
Figure 1.9). On the other hand, words in the product advertisements (Figure 1.10) from
one company are scattered all over the map, and we cannot separate each area by each
company. Finally, we find from Figure 1.11 that referee comments restore the hidden
clusters of products from companies. The value of isolation degree is not as large as in
Figure 1.9, but larger than in Figure 1.10. (See Iso values in the figure captions.)
We conducted similar experiments for the data on the design competition from
2003, 2004, and 2005. That is, we have three sets of {data a, data b, data c} for the cor-
responding 3 years. For all 3 years, we obtained similar results, as reported above. The
isolation degree was largest for Data a, the least for Data b, and the second for Data c.
Thus, we can conclude that the designers in the same company share their con-
cepts for design, and these concepts tend to be hidden to people in the market. This
occurs in a product path starting from the designers all the way to the marketers/
advertisers, because of the weak impression of designers’ opinions to other mem-
bers in the company. Referees do communicate among themselves when deciding
to award the best products, with some sort of design expertise that is common to
concepts in the minds of designers.
1.7 Conclusions
In the two cases presented, we have shown the collaborations by marketers and design-
ers of new products. These collaborations were executed based on scenario maps cre-
ated by KeyGraph for the data on sales and patents. By following the well-organized
C7019.indb 15 12/18/08 11:33:11 AM

communication process for sensing novel values in the market, the participants suc-
cessfully developed hit products.
However, collaborations in a company do not always work, especially if partici-
pants come from different working sections. We analyzed the thoughts of designers
and the people surrounding designers. Pictorial KeyGraph has been introduced for
this comparison. The ideas originally shared by the designers belonging to the same
company can be clearly expressed by words, even though they might be engaged
daily in different projects for creating different products. However, these design-
ers’ ideas tended to be hidden in the products’ path from the marketers/advertisers
because of the weak impression of the designers’ communication with other mem-
bers of the company.
Thus we should consider the difference of inter- and intrasection communications
in design and marketing and invent methods for aiding integration of these com-
munications into creating business strategies of a company. The belief underlying
this book is that tools of data mining and their well-organized coupling with human
communication have the possibility to achieve this goal. The reader is referred to
articles on the same project by heterogeneous members for creative design, such as
Eris and Fruchter.18,19 We are encouraged also by the literature of cognitive sciences
for design, which says ambiguous information can trigger design creations.20 All in
all, information from other teams, which might be ambiguous, may motivate creative
designs with rich originality, although it may not be as familiar and well understood
as messages from collaborators. Data mining and visualization play an important
role to provide a scenario map on which collaborators can exchange ambiguous but
useful information.
References
1. Fader, P.S. and Hardie, B.G.S. Forecasting repeat sales at CDNOW: a case study.
The Marketing Department, Working Paper No. 99-023. Wharton, Philadelphia, PA,
1999.
2. Prasun, D. and Subhasis, C. (2007) Prediction of retail sales of footwear using feed-
forward and recurrent neural networks. Neural Comp Appl. 16, 491–502.
3. Agrawal, R., Imielinski, T., and Swami, A. (1993) Mining association rules between
sets of items in large data. Proc ACM SIGMOD. 207–216.
4. Yamashita, T., Hirasawa, K., and Hu, J. (2005) Multi-branch neural networks and its
application to stock price prediction, in Proc KES Conf. Springer, LNCS3681.
5. Weiss, G.M. and Hirsh, H. Learning to predict rare events in event sequences, in Proc
Fourth Int Conf Knowledge Discovery Data Mining (KDD-98). AAAI Press, Menlo
Park, CA, 1998, 359–363.
6. Hirooka, Y., Terano, T., and Otsuka, Y. Extending content-based recommendation by
order-matching and cross-matching method, in Proc Electron Comm Web Technol:
First Int Conf. Springer, LNCS1875, 177–190, 2000.
7. Donaldson, M. Human Minds. AAllen Lane, 1993.
8. Ohsawa, Y., (2003). KeyGraph: visualized structure among event clusters, in Chance
Discovery, Ohsawa, Y. and McBurney, P., eds., Springer-Verlag, 2003, 262–275.
9. Ohsawa, Y., Nels, E.B., and Yachida, M. (1998) KeyGraph: automatic indexing by co-
occurrence graph based on building construction metaphor. Proc Adv Digital Library
Conf (IEEE ADL’98), 12–18.
C7019.indb 16 12/18/08 11:33:11 AM

10. Okazaki, N. and Ohsawa, Y. Polaris: an integrated data miner for chance discovery, in
Proc Third Int Workshop Chance Discovery Manage. Greece, 2003.
11. Ohsawa, Y. and Usui, M. Creative marketing as application of chance discovery, in
Chance Discoveries in Real World Decision Making, Ohsawa, Y. and Tsumoto, S.,
eds., Springer, Berlin, 2006.
12. Horie, K. and Ohsawa, Y. (2006) Product designed on scenario maps using pictorial
KeyGraph. WSEAS Trans Information Sci Application. 3:1324–1331.
13. Horie, K., Yoshiharu, M., and Yukio, O. (2007) Human-interactive annealing process
with pictogram for extracting new scenarios for patent technology. Data Sci J. 6,
S132–S136.
14. Ohsawa, Y. and Nara, Y. (2003) Understanding internet users on double helical model
of chance-discovery process. J New Generation Comp., 21, 109–122.
15. Suwa, M. Meta-cognitive verbalization as a tool for acquiring expertise through differ-
entiation and problem-framing. Poster Abstracts Vol 2nd Int Conf Design Computing
and Cognition (DCC’06). 43–44, 2006.
16. Chen, H. Approach on the vocabulary problem in collaboration. In: Proceedings
National Science Foundation Coordination Theory and Collaboration Technology
Workshop, Arlington, VA, 1993.
17. Good design finder: a library of good design award (2005). http://www.g-mark.org/
search/index.jsp
18. Eris, O. Effective Inquiry for Innovative Engineering Design. Kluwer Academic, MA,
2004.
19. Fruchter, R., Ohsawa, Y., and Matsumura, N. (2005) Knowledge reuse through chance
discovery from an enterprise design-build enterprise data store. New Math Natural
Comp. 3, 393–406.
20. Gaver W.W., Beaver, J., and Benford, S. (2003) Ambiguity as a resource for design.
Proc Comp Human Inter., pp. 233–240.
C7019.indb 17 12/18/08 11:33:11 AM

2 Reframing the
Data-Mining Process
David Bergner and Ozgur Eris
Contents
2.1 Introduction...................................................................................................... 19
2.2 The Data-Mining Process Reframed...............................................................20
2.3 Decision Quality.............................................................................................. 22
2.3.1 The Decision Basis............................................................................... 23
2.3.2 The Decision Frame..............................................................................24
2.3.3 How Process Affects Quality—Comprehension of Prospects.............24
2.4 Measuring Team Interaction Processes...........................................................26
2.4.1 A Measure of Framing..........................................................................26
2.4.2 A Measure of Inquiry........................................................................... 27
2.4.3 Prior Results..........................................................................................28
2.5 Discussion........................................................................................................ 29
2.6 Implications...................................................................................................... 30
2.6.1 Team Composition................................................................................ 30
2.6.2 Team Interaction................................................................................... 31
2.6.3 Decision Focus...................................................................................... 31
2.7 Limitations....................................................................................................... 32
2.8 Conclusions...................................................................................................... 32
References................................................................................................................. 33
2.1 Introduction
Imagine that two different data-mining teams of apparently comparable makeup and
competence were given the same tools, access to the same data sets, and equal time to
work. Later, the business managers who sponsored the projects looked back on their
outcomes, and observed that one team’s results were highly valued and influenced
the business, whereas the other team’s results were insignificant and soon forgot-
ten. Imagine the managers now seeking to determine why two apparently identi-
cal projects could produce such different results. How would we help them? How
would we study the problem? What performance variables would we measure? One
way to address these questions is to consider how data mining can shape business
decisions.
19
C7019.indb 19 12/18/08 11:33:11 AM

In this chapter, we extend prior models of the human role in data mining. We
build on prior literature regarding decision-making, design, and team interaction to
model relationships of data-mining processes to business decisions. Based on this
model, we suggest ways to measure and manage the data-mining process to enhance
business value creation. We reframe the data-mining process to further emphasize its
social and organizational, rather than computational aspects, and to focus explicitly
on how the process can improve decision quality.
The notions of frames, framing, and reframing are central to our discussion of
how interaction processes shape decisions. A frame establishes context and deter-
mines what is considered part of an activity, problem, or decision; it determines what
is worthy of attention and discussion and what is not. Framing and reframing refer
to the processes by which individuals and teams establish and shift their frames.
Although frames have substantial implicit aspects, here we are concerned with their
explicit aspects, which are revealed in communication, and provide a basis for empir-
ical investigation and analysis of frames (Bergner, 2004).
This chapter’s title, “Reframing the Data-Mining Process,” has two distinct mean-
ings. First, it refers to shifts in the way we as theorists and practitioners conceive
of data mining—what aspects of the data-mining process will receive our attention.
Second, it refers to ongoing shifts in the way data-mining teams and their sponsors
conceive of their projects—what aspects of the data and the business will receive their
attention. In the following sections we deal with these two topics, and then consider
how to measure and manage data-mining team processes to create business value.
2.2 The Data-Mining Process Reframed

Fayadd, Shapiro, and Smith (1996) showed that successful knowledge discovery
from data mining hinges not only on the data sets and data-mining tools available,
but perhaps even more so on human interpretation and choices that shape the data-
mining process and results. Ohsawa (2002) built on this model to provide an even
richer description of the human role in observing, interpreting, selecting, and evalu-
ating data and explaining the significance of discoveries in terms of decision mak-
ing. Considering factors beyond the discovery of knowledge per se, Ohsawa was
concerned with the discovery of chance (Ohsawa & McBurney, 2003)—a chance
being an event or situation having significant impact on decision making or prob-
lem solving, and discovery of chance entailing both learning of such an event or
situation and explaining its significance for decision making or problem solving.
Drawing on insights from decision theory and decision analysis practice, Bergner
(2004) elaborated on Ohsawa’s model by focusing on the second criterion for dis-
covery of chance—explaining how the discovery is significant for decision making.
This could provide a test of the value created by a data-mining process—we would
ask what decisions were influenced by the process outcomes, what was the nature of
the influence, and how important were the decisions. Projects applying data mining
for design and marketing, given their business context, are concerned with discovery
of chance to the extent project sponsors will expect actionable results from the pro-
cess (actionable: capable of being acted on [Merriam-Webster OnLine]; relating to
C7019.indb 20 12/18/08 11:33:11 AM

Reframing the Data-Mining Process 21
or being information that allows a decision to be made or action to be taken [www.

answers.com]).
Ohsawa (2002) also introduced the concept of subject-data—data generated by
the team members as they interact—in contrast to object-data, which are the data
the team intends to mine. The term object-data simply reflects that the data set is the
object of the data-mining activity. Subject-data reflects knowledge, perspectives, and
interpretations of the team members themselves, which emerge in communication
and shape the course of the data-mining process. Subject data include data about
the team members (such as who talks the most, who has expertise in a given field),
as well as data they explicitly supply about the object-data or about the world in
general. Such data would be found in records and artifacts of team interaction, such
as documents, presentations, and even marker board notes, or, given a systematic
knowledge capture process, perhaps in videotapes, audiotapes, transcripts of team
interaction, or notes taken by trained observers. In the next section, we suggest ways
to collect and interpret subject-data, but first we consider how the foregoing literature
reframes the data-mining process.
Figure 2.1 shows two frames for the data-mining process, Frame-1 and Frame-2.
Because Frame-2 contains Frame-1, Frame-2 represents a broader perspective.
Process1 and Process2 are subparts of the entire data mining process considered in
this chapter. In general, our notion of process entails, for example, team interactions;
tools, techniques, and procedures employed; starting and ending criteria; inputs; and
results.
Frame-1 focuses on the computational aspects we typically attribute to data min-
ing: a team with technical expertise employs particular tools and algorithms to mine
some data set (the object-data), and somehow, the results should generate knowledge
relevant to business decisions. From the perspective of Frame-1, the quality of the
outcome hinges on qualities of the data set, the data-mining technologies and tech-
niques employed, and the technical expertise of the team. Frame-1 considers busi-
ness decisions, but decisions are more central to Frame-2.
Frame-2 focuses on social aspects emphasized in the literature cited previously:
team members with unique roles, expertise, and perspectives generate subject-data
through communication as they make interpretations and interact in an organiza-
tional setting to accomplish Process1. Process2 captures and mines these subject-data
and shapes Process1. In addition, Process2 elicits knowledge relevant to business
Frame-1–Computational focus Frame-2–Social focus

Generates Decision- Elicits
relevant
Frame-3 knowledge
Object- Mines Generates Subject- Mines

Process1 Process2
data data
Shapes
Figure 2.1 Data-centric decision-design.
C7019.indb 21 12/18/08 11:33:12 AM

decisions, which may be formally elicited from business decision makers or emerge
dynamically based on subject data generated by Process1. In this sense, to elicit means
to draw out and structure. The structure of decision-relevant knowledge is discussed in
Section 2.3.1, “The Decision Basis,” and Section 2.3.2, “The Decision Frame.”
The bold arrows in Figure 2.1 depict key information flows by which business
decisions both inform and are informed by the data mining process. Accumulation
and interpretation of subject-data are central to these information flows. From the
perspective of Frame-2, the quality of the outcome hinges on qualities of team inter-
action and interpretation, and team interaction with business decision makers is
critical to inform the data-mining process and to communicate insights that shape
decisions. We call this “data-centric decision-design” to emphasize the overall goal
of the activity—to create decision opportunities for business decision makers based
on synthesis of insights obtained from the object data and the team members.
We do not wish to imply that Frame-1 and Frame-2, as we have described them,
constitute the entirety of frames and framing processes relevant to data mining. For
example, we could consider another frame, Frame-3, around the object-data itself,
partially overlapping Frame-1 and Frame-2 but also including unknown computa-
tional, organizational, and subjective factors that influenced processes by which the
object-data were created and came to be the object of a data-mining project. Or we
could consider Frame-0 (not depicted in Figure 2.1), a broader frame that would
subsume Frame-2 while including additional factors, such as aspects of the organi-
zational decision-making practices, setting, and cultural context in which the entire
process is carried out. Regarding this broader frame, we assume data-mining projects
are undertaken by teams with sponsors who have the power to commit resources and
may act on decision-relevant knowledge that is generated. Further modeling of busi-
ness decision-making practices is beyond our scope. Focusing on Frame-2 allows us
to carve out a manageable piece of the overall framing problem.
To varying degrees, the fundamental concerns of Frame-2 would be reflected in
all successful data-mining processes. However, Frame-2 places novel emphasis on
purposefully capturing and interpreting subject-data to shape a dynamic and con-
textually richer iterative process and systematically eliciting knowledge relevant to
business decisions. This emphasis motivates us to consider additional tools, tech-
niques, and procedures suitable for eliciting, capturing, and mining subject-data and
modeling decisions and also motivates new perspectives on team composition. Such
an approach would depend on team interaction models, as well as a rich set of dis-
tinctions regarding decisions and decision-making processes. We consider decision
making next, and then team interaction.
2.3 Decision Quality

This discussion of decision quality draws on Bergner (2006), Howard (1988), and
McNamee and Celona (2001). We assume the business context is challenging (i.e.,
having aspects such as uncertainty, complexity, a changing strategic environment,
new situations, long time horizons, or lack of established precedents. This appears
congruent with contexts where data mining would be applied for design and market-
ing. Decision making in challenging contexts requires careful thought, ingenuity,
C7019.indb 22 12/18/08 11:33:12 AM

Another Random Document on
Scribd Without Any Related Topics
To doze just with the whistle of that old 10.30 train.
Ah, life is not of solitude,

Nor childhood joys alone,
Its mirth not all departed, though
We reap the evil sown.
But nights of rain and solitude
Bring back the happy past—
The freight that came so regular
My eyes to close at last.
From all the now I quick would flee—
It seems so full of pain—
If I could sleep forever with that whistle’s wail again!
Gallows Gate
BEING AN INCIDENT IN THE LIFE OF DICK RYDER, OTHERWISE
GALLOPING DICK, SOMETIME GENTLEMAN OF THE ROAD
BY H. B. MARRIOTT-WATSON
’T
WAS two o’clock of a bright wild March day that I cleared St.
Leonard’s Forest, and came out upon the roads at the back of
Horsham. I was for Reading, but chose that way by reason of the
better security it promised, which, as it chanced, was a significant
piece of irony. Horsham, a mighty quiet, pretty town, lay in a blaze
of the sun, enduring the sallies of a dusty wind, and, feeling hot and
athirst after my long ride, I pulled up at an inn and dismounted.
“Host,” says I, when I was come it; “a pint of your best Burgundy
or Canary to wash this dust adown; and rip me if I will not have it
laced with brandy.”
“Why, sir,” says he, “a cold bright day for horseback,” and shakes
his head.
“Damme, you’re right,” says I. “Cold i’ the belly and hot in the
groin. Here’s luck to the house, man,” and I tossed off the gallipot.
“Why, goodman, ye’ll make your fortune on this,” I said with a
derisive laugh, and flung open the door, to go out; when all of a
sudden I came to silence and a pause.
“’Tis the officers,” says the landlord, who was at my ear.
“Gadslife, ’tis the sheriff’s men from Lewes.”
“Lewes!” says I slowly; “what be they here for?”
“Why,” says he in a flutter, “there was him that was taken for a
tobyman by Guilford. He was tried at Lewes, and will hang.”
“If he be fool enough to be taken, let him be hanged and be
damned,” says I carelessly.
When I was got upon my horse I began to go at a walk down the
High street, for though, as was according to nature, I was inquisitive
about the matter, I was too wary to adventure ere I was sure of my
ground. And this denial of unnecessary hazards, as is my custom,
saved me from a mishap; for as the procession wound along, the
traps and the carriage between ’em, there was one of them that
turned his head aside to give an order, and, rip me if ’twas not that
muckworm, traitor and canter, the thief-taker, Timothy Grubbe. I had
an old score with Timothy, the which I had sworn to pay; but that
was not the time nor the opportunity, and so I pulled in and lowered
my head, lest by chance his evil eye might go my way. As I did so,
something struck on the mare’s rump, and, looking back, I saw a
young man on horseback that had emerged from a side street.
“Whoa, there,” says I cheerfully. “Are you so blinded by March
dust as not to see a gentleman when he goes by?”
He was a slight, handsome-looking youth, of a frank face but of a
rustic appearance, and he stammered out an apology.
“Why, I did but jest,” I said heartily. “Think no more on’t,
particularly as ’twas my fault to have checked the mare of a sudden.
But to say the truth I was gaping at the grand folks yonder.”
He stared, after the traps and says he in an interested voice:
“Who be they? Is it my Lord Blackdown?”
Now this comparison of that wry-necked, pock-faced villain
Grubbe to a person of quality tickled me, but I answered, keeping a
straight face.
“Well, not exactly,” says I, “not my lord, but another that should
stand or hang as high maybe, and shall some day.”
“Oh,” says he, gazing at me, “a friend of yours, sir?”
He was a ruddy color, and his mouth was habitually a little open,
giving him an expression of perpetual wonder and innocence; so
that, bless you, I knew him at once for what he was at heart—a
simple fellow of a natural kindliness and one of no experience in the
world, and a pretty dull wit.
“Not as you might call him, a friend,” said I gravely, “but rather
one that has put an affront upon me.”
“You should wipe it out, sir,” says this innocent seriously. “I would
allow no man to put an affront on me, gad, I would not!”
“Why,” said I drily, “I bide my time, being, if I may say so, of less
mustard and pepper than yourself. Nevertheless it shall be wiped out
to the last stain.”
“Gad, I like that spirit,” says he briskly, and, as if it constituted a
bond betwixt us, he began to amble slowly at my side. “If there is
any mischief, sir,” says he, “I trust you will allow me to stand your
friend.”
Here was innocence indeed, yet I could ha’ clapped him on the
back for a buck of good-fellowship and friendliness, and, relaxing my
tone, I turned the talk on himself.
“You are for a journey?” says I.
He nodded, and his color rose, but he frowned. “I am for
Effingham,” he answered.
“So am I,” said I, “at least I pass that way,” which was not so, for
I was for Reading, and had meant to go by Guilford. Yet I was in no
mind to risk an encounter with Grubbe and his lambs, who were
bound for Guilford if what the innkeeper said was true; and the way
by Effingham would serve me as well as another.
He looked pleased, and says he: “Why, we will travel in
company.”
“With all my heart!”
The traps had disappeared upon the Guilford road in a mist of
dust, and we jogged on comfortably till we came to cross-roads,
where we turned away for Slinfold, reaching that village near by two
of the clock. Here my companion must slake his thirst, and I was
nothing loath. He had a gentlemanly air about him for all his rustic
habit, and very pleasantly, if with some awkwardness, offered me of
a bottle.
“You mind me,” said I, drinking to him, for I liked the fellow, “of a
lad that I knew that was in the wars.”
“Was you in the wars?” asks he eagerly.
I had meant the wars of the road, which, indeed, are as perilous
and as venturesome as the high quarrels of ravening nations.
“I served in Flanders,” said I.
“My father fought for His Gracious Majesty Charles I,” says he
quickly, “and took a deep wound at Marston Moor. There was never
a braver man than Squire Masters of Rockham.”
“I’ll warrant his son is like him,” said I.
He bowed as if he were at Court. “Your servant, sir,” says he,
smiling well pleased, and eyed me. “You have seen some service,
sir?”
“Why, as much as will serve, Mr. Masters.”
He looked at me shyly. “You have my name, now?” said he, and
left his question in the air.
“You may call me Ryder,” said I.
“You have had your company?” he went on in a hesitating voice.
“Not always as good company as this,” I replied, laughing.
“I knew it,” he said eagerly; “you are Captain Ryder?”
“There have been those that have put that style on me,” I
answered, amused at his persistence.
“I am glad that I have met you, Captain,” said this young fool,
and put his arm in mine quite affectionately.
“I have been unhappily kept much at home, and have seen less
than I might of things beyond the hills. Not but what Sussex is a fine
shire,” he adds, with a sigh.
“Why, it is fine if so be your home be there,” I replied.
“My home is there,” he said, and paused, and again the frown
wrinkled up his brow.
He said no more till we were in the saddle again and had gone
some half a mile, and then he spoke, and I knew his poor brain had
been playing pitch and toss with some thought.
“Captain Ryder,” he said abruptly, “you have traveled far and seen
much. You might advise one junior to you on a matter of worldly
wisdom.”
Sink me, thinks I, what’s the boy after? But, says I gravely, from
a mutinous face: “You can hang your faith on me for an opinion or a
blow, Mr. Masters.”
“Thank you,” says he heartily, and then thrust a hand into his
bosom and rapidly stuck at me a document. “Read that, sir,” said he
impulsively.
I opened it, and found ’twas writ in a woman’s hand, and
subscribed Anne Varley; and the marrow of it was fond affection.
Why, ’twas but a common love billet he had given me, of the
which I have seen dozens and received very many—some from
persons of quality that would astonish you. But what had I to do
with this honest ninny and his mistress? I had no nose for it, and so
said I, handing him back his letter.
“It has a sweet smack and ’tis pretty enough inditing.”
“Ah,” says he quickly, “’tis her nature, Captain. ’Tis her heart that
speaks. Yet is she denied by her parents. They will have none of
me.”
“The more to their shame,” I said.
“They aspire high,” says he, “as Anne’s beauty and virtues of
themselves would justify. Yet she does love me, and I her, and we
are of one spirit and heart. See you how she loves me, poor thing,
poor silly puss! And they would persuade her to renunciation. But
she shall not—she shall not; I swear it!” he cried in excitement. “She
shall be free to choose where she will.”
“Spoke like a man of temper,” said I approvingly. “You will go win
her forthright.”
“I am on my journey to accomplish that now,” says he. “She has
writ in this letter, as you have seen, that her father dissuades her,
and she sighs her renunciation, adding sweet words of comfort that
her affection will not die—no, never, never, and that she will die
virgin for me. Say you not, sir, that this is beautiful conduct, and, am
I not right to ride forth and seize her from her unnatural parents, to
make her mine?”
“Young gentleman,” said I, being stirred by his honest sincerity
and his bubbling over, “were you brother to me, or I to Mistress
Anne, you should have my blessing.”
At that he glowed, and his spirits having risen with this
communication, he babbled on the road of many things cheerfully,
but mostly of love and beauty, and the virtues of Mistress Anne of
Effingham Manor.
I will confess that after a time his prattle wearied me; ’twas too
much honey and cloyed my palate. If he had known as much of the
sex as has fallen to my lot he would have taken another stand, and
sung in a lower key.
Well, ’twas late in the afternoon when we reached the hills
beyond Ewhurst, and began to climb the rugged way to the top. The
wind had gone down with the sun in a flurry of gold in the west, to
which that eastern breeze had beat all day; and over the head of
Pitch Hill last year’s heather still blazed in its decay.
When we had got to the Windmill Inn, that lies packed into the
side of the wooded hill, we descended for refreshment, and I saw
the horses stalled below for baiting. Now that house, little and quiet,
perches in a lonely way in the pass of the hill, and upon one side the
ground falls so fast away that the eye carries over a precipitous
descent toward the weald of Surrey and the dim hills by the sea.
And this view was fading swiftly in the window under a bleak sky as
Masters and I ate of our dinner in the upstairs room that looked
upon it. He had a natural grace of mind despite the rawness of his
behavior, and his sentiments emerged sometimes in a gush, as
when, says he, looking at the darkening weald:
“I love it, Captain. ’Tis mine. My home is there, and, God willing,
Anne’s too shall be.”
“Amen!” said I heartily, for the boy had gone to my heart, absurd
though he was.
And just on that there was a noise without the door, the clank of
heavy feet rang on the boards, and Timothy Grubbe’s ugly mask
disfigured the room.
He came forward a little with a grin on his distorted features,
and, looking from one to the other of us, said he:
“My respects, Captain, and to this young plover that no doubt
ye’re plucking. By the Lord, Dick Ryder, but I had given you up.
Heaven sends us good fortune when we’re least thinking of it.”
Masters, at his word, had started up. “Who are you, sir, that
intrudes on two gentlemen?” he demanded with spirit. “I’ll have you
know this is a private room. Get you gone!”
“Softly, man,” says Grubbe, in an insinuating voice. “Maybe I’m
wrong and you’re two of a color. Is it an apprentice, Dick, this brave
lad that talks so bold and has such fine feathers?”
“If you do not quit,” said I shortly, “I will spit your beauty for you
in two ticks.”
“Dick Ryder had always plenty of heart,” said he in his jeering
way. “Dick had always a famous wit, and was known as a hospitable
host. So I will take the liberty to invite to his sociable board some
good fellows that are below, to make merry. We shall prove an
excellent company, I’ll warrant.”
Masters took a step toward him.
“Now, who the devil soever you may be, you shall not use
gentlemen so,” he cried, whipping out his blade.
But Grubbe turned on him satirically. “As for you, young
cockchafer,” said he, “it bodes no good to find you in this company.
But as you seem simpleton enough, I’ll give you five minutes to take
your leave of this gentleman of the road. Dick, you’re a fine
tobyman, and you have enjoyed a brave career, but, damme, your
hour is struck.”
I rose, but, ere I could get to him, young Masters had fallen on
him.
“Defend yourself, damn ye,” he said, “you that insult a gentleman
that is my friend! Put up your blade!” and he made at him with
incredible energy.
Uttering a curse Grubbe thrust out his point and took the first
onrush, swerving it aside; and ere I could intervene they were at it.
My young friend was impetuous, and as I saw at once, none too
skilful; and Grubbe kept his temper, as he always did. He stood with
a thin, ugly smile pushing aside his opponent’s blade for a moment
or two, until, of a sudden, he drew himself up and let drive very low
and under the other’s guard. The sword rattled from Masters’s hand,
and he went down on the floor. I uttered an oath.
“By God, for this shall you die, you swine!” said I fiercely; and I
ran at him; but, being by the door, he swept it open with a
movement and backed into the passage.
“The boot is on t’other leg, Dick,” says he maliciously. “’Tis you
are doomed!” and closing the door behind him he whistled shrilly.
I knew what he intended, and that his men were there, but I
stooped over the boy’s body and held my fingers to his heart. ’Twas
dead and still. I cursed Grubbe and started up. If I was not to be
taken there was only the window, looking on the deeps of the
descending valley. I threw back the casement and leaped over the
sill. Grubbe should perish, I swore, and I doubled now my oath.
I could ha’ wept for that poor youth that had died to avenge my
honor. But my first business was my safety, and I crept down as far
as I might and dropped. By that time the catchpolls were crowding
into the room above. I struck the slanting hill and fell backward, but,
getting to my feet, which were very numb with the concussion of the
fall, I sped briskly into the darkness, making for the woods.
I lay in their shelter an hour, and then resolved on a
circumspection. ’Twas not my intention to leave the mare behind, if
so be she had escaped Grubbe and his creatures; and, moreover, I
had other designs in my head. So I made my way back deviously to
the inn and reconnoitered. Stillness hung about it, and after a time I
marched up to the door cautiously and knocked on it.
The innkeeper opened it, and, the lamp burning on my face,
started as if I were the devil.
“Hush, man!” said I. “Is the officer gone?”
He looked at me dubiously and trembling. “Come,” said I, for I
knew the reputation of those parts, “I am from Shoreham Gap
yonder, and I was near taken for an offense against the revenue.”
“You are a smuggler?” said he anxiously. “They said you were a
tobyman.”
“They will take away any decent man’s name,” said I. “I want my
horse. You have no fancy for preventive men, I’ll guess.”
And this was true enough, for he had a mine of cellars under his
inn and through the roadway.
“But your friend?” said he, still wavering. “Him that is dead——”
“As good a man as ever rolled a barrel,” said I.
He relaxed his grip of the door. “’Tis a sore business for me this
night,” he complained.
“Nay,” said I. “For I will rid your premises of myself and friend, by
your leave, or without it,” says I.
He seemed relieved at that, and I entered. The horses were safe,
as I discovered, for Grubbe must have been too full of his own prime
business to make search, and, getting them out, I made my
preparations. I strapped the lad’s body in the stirrups, so that he lay
forward on the horse with his head a-wagging; but (God deliver
him!) his soul at rest. And presently we were on the road, and
threading the wilderness of the black pine woods for the vale below
toward London.
The moon was a glimmering arc across the Hurtwood as I came
out on the back of Shere, and, pulling out of the long lane that gave
entry to the village, reined up by the “White Horse.” From the inn
streamed a clamor of laughter, and without the doorway and
wellnigh blocking it was drawn up a carriage with a coachman on his
seat that struck my eyes dimly in the small light. I was not for calling
eyes on me with a dead man astride his horse, so I moved into the
yard, thinking to drain a tankard of ale if no better, before I took the
road over the downs to Effingham. But I was scarce turned into the
yard ere a light flaring through the window poured on a face that
changed all the notions in my skull. ’Twas Grubbe!
Leaving the horses by I returned to the front of the inn, and says
I to the coachman that waited there, as I rapped loud on the door:
“’Tis shrewish tonight.”
“Aye,” says he in a grumbling, surly voice. “I would the country
were in hell.”
“Why, so ’twill be in good time,” said I cheerfully; and then to the
man that came, “Fetch me two quarts well laced with gin,” says I,
“for to keep the chill of the night and the fear o’ death out.”
The coachman laughed a little shortly, for he knew that this was
his invitation.
“Whence come you then?” said I, delivering him the pot that was
fetched out.
He threw an arm out. “Lewes,” said he, “under charge with a
tobyman that was for chains yonder.”
He nodded toward the downs and drank. I cast my eyes up and
the loom of the hill just t’other side of the village was black and
ominous.
“Oh,” says I, “he hangs there?”
“At the top of London Road,” says he, dipping his nose again.
“There stands the gallows, where the roads cross and near the
Gate.”
“Gallows Gate,” said I, laughing. “Well, ’twas a merry job
enough.”
“Aye,” says he. “But by this we might ha’ been far toward London
Town, whither most of us are already gone. But ’twas not his wish.
He must come back with the Lewes sheriff and drink him farewell.”
“Leaving a poor likely young man such as yourself to starve of
cold and a empty belly here,” said I. “Well, I would learn such a one
manners in your place, and you shall have another tankard of dogs-
nose for your pains,” says I, whereat I called out the innkeeper
again, but took care that he had my share of the gin in addition to
his own. By that time he was garrulous, and had lost his caution, so,
keeping him in talk a little and dragging his wits along from point to
point, I presently called to him.
“Come down,” said I, “and stamp your feet. ’Twill warm you
without as the liquor within.” And he did as I had suggested without
demur.
“Run round to the back,” says I, “and get yourself a noggin, and
if so be you see a gentleman on horseback there asleep, why, ’tis
only a friend of mine that is weary of his long journey. I will call you
if there be occasion.”
He hesitated a moment, but I set a crown on his palm, and his
scruples vanished. He limped into the darkness.
’Twas no more than two minutes later that I heard voices in the
doorway, and next came Timothy Grubbe into the night, in talk with
someone. At which it took me but thirty seconds to whip me into the
seat and pull the coachman’s cloak about me, so that I sat stark and
black in the starlight. Grubbe left the man he talked with and came
forward.
“You shall drink when ye reach Cobham, Crossway,” says he,
looking up at me, “and mind your ways, damn ye!”
And at that he made no more ado, but humming an air he
lurched into the carriage. I pulled out the nags, and turned their
heads so that they were set for the north. And then I whistled low
and short—a whistle I knew that the mare would heed, and I trusted
that she would bring her companion with her. The wheels rolled out
upon the road and Timothy Grubbe and I were bound for London all
alone.
As I turned up the London road that swept steeply up the downs
I looked back, and behind the moon shone faintly on Calypso and
behind her on the dead man wagging awkwardly in his stirrups.
I pushed the horses on as fast as might be, but the ruts were still
deep in mud, and the carriage jolted and rocked and swayed as we
went. The wind came now with a little moaning sound from the
bottom of the valley, and the naked branches creaked above my
head, for that way was sunken and tangled with the thickets of nut
and yew. And presently I was forced to go at a foot pace, so abrupt
was the height. The moon struck through the trees and peered on
us, and Grubbe put his head forth of the window.
“Why go you not faster, damn ye?” says he, being much in liquor.
“’Tis the hill, your honor,” said I.
He glanced up and down.
“What is it comes up behind?” says he, shouting. “There is a
noise of horses that pounds upon the road.”
“’Tis the wind,” says I, “that comes off the valley and makes play
among the branches.”
He sank back in his seat, and we went forward slowly. But he
was presently out again, screaming on the night.
“There is a horseman behind,” says he. “What does he there?”
“’Tis a traveler, your honor,” says I, “that goes, no doubt, by our
road, and is bound for London.”
“He shall be bound for hell,” says he tipsily, and falls back again.
The horses wound up foot by foot and emerged now into a space
of better light, and I looked around, and there was Grubbe, with his
head through the window and his eyes cast backward.
“What fool is this,” says he, “that rides so awkwardly, and drives
a spare horse? If he ride no better, I will ask him to keep me
company, if he be a gentleman. Many gentlemen have rode along of
me, and have rode to the gallows tree,” and he chuckled harshly.
“Maybe he will ride with you to the Gallows Gate, sir,” says I.
“Why, Crossway,” says he, laughing loudly, “you have turned a
wit,” and once more withdrew his head.
But now we were nigh to the top of the down, and I could see
the faint shadow of the triple beam. With that I knew my journey
was done, and that my work must be accomplished. I pulled to the
horses on the rise, and got down from my seat.
“Why d’ye stop, rascal?” called Grubbe in a fury, but I was by the
door and had it opened.
“Timothy Grubbe,” said I, “ye’re a damned rogue that the devil,
your master, wants and he shall have ye.”
He stared at me in a maze, his nostrils working, and then says he
in a low voice: “So, ’tis you.”
“Your time has come, Timothy,” said I, flinging off my cloak, and
I took my sword. “Out with you, worm.”
He said never a word, but stepped forth, and looked about him.
He was sobered now, as I could see from his face, which had a
strange look on it.
“Ye’re two rascals to one, Dick,” says he slowly, looking on the
dead man on his horse which had come to a stop in the shadows.
“No,” says I, “this gentleman will see fair play for us.”
Grubbe took a step backward. “Sir,” says he, addressing the dead
man—but at that moment Calypso and her companion started, and
came into the open.
The moon shone on the face of the dead. Grubbe uttered a cry,
and turned on me. His teeth showed in a grin.
“No ghost shall haunt me, Dick,” says he. “Rather shall another
ghost keep him company,” and his wry neck moved horribly.
I pointed upward where the tobyman hung in chains, keeping his
flocks by moonlight. “There’s your destiny,” said I. “There’s your
doom. Now defend, damn ye, for I’ll not prick an adder at a
disadvantage.”
He drew his blade, for no man could say that Timothy Grubbe,
time-server and traitor as he was, lacked courage. Suddenly he
sliced at me, but I put out and turned off the blow.
“If you will have it so soon,” said I, “in God’s name have it,” and I
ran upon him.
My third stroke went under his guard, and I took him in the
midriff. He gave vent to an oath, cursed me in a torrent, and struck
at me weakly as he went down.
He was as dead as mutton almost ere he touched the ground.
I have never been a man of the church, nor do I lay any claim to
own more religion than such as to make shift by when it comes to
the end. No, nor do I deny that I have sundry offenses on my
conscience, some of which I have narrated in my memoirs. But
when it comes to a reckoning I will make bold to claim credit in that
I rid the world he had encumbered of Timothy Grubbe—the foulest
ruffian that ever I did encounter in the length of my days on the
road.
I climbed the beam and lowered the poor tobyman, and it took
me but a little time to make the change. The one I left where he had
paid the quittance in the peace of the earth, and t’other a-swinging
under the light of the moon on Gallows Gate.
I have said my journey was done, but that was not so. There was
more for me to do, which was to deliver poor Masters at his lady-
love’s and break the unhappy news. And so, leaving the carriage
where it stood with the patient horses that were cropping the grass,
I mounted the mare and began to go down the long limb of the
downs to the north.
’Twas late—near midnight—when I reached Effingham and found
my way to the manor. I rapped on the door, leaving Calypso and
t’other in the shadows of the house, and presently one answered to
my knock.
“What is it?” says she.
“’Tis a stranger,” says I, “that has news of grave import for
Mistress Anne Varley, whom I beg you will call.”
“She cannot hear you,” said she. “’Tis her wedding night.”
“What!” said I in amazement, and instantly there flowed in on me
the meaning of this.
“Curse all women save one or two!” thinks I. And I turned to the
maid again with my mind made up.
“Look you, wench,” said I. “This is urgent. I have an instant
message that presses. And if so be your mistress will bear with me a
moment and hold discourse, I’ll warrant she shall not regret it—nor
you,” says I, with a crown piece in my palm.
She hesitated and then, “Maybe she will refuse,” says she. “She
hath but these few hours been wed.”
“Not she,” said I, “if you will tell her that I bring good news,
great news—news that will ease her spirit and send her to her bridal
bed with a happy heart.”
At that she seemed to assent, and with my crown in her hand
she disappeared into the darkening of the house. It must have been
some ten minutes later that a light flashed in the hall and a voice
called to me.
“Who is it?” it asked, “and what want you at this hour?”
I looked at her. She was of a pretty face enough, rather pale of
color, and with eyes that moved restlessly and measured all things.
Lord, I have known women all my life in all stations, and I would ha’
pinned no certainty on those treacherous eyes. She was young, too,
but had an air of satisfaction in herself, and was in no wise
embarrassed by this interview. I had no mercy on her, with her oaths
of constancy writ in water that figured to be tears and her false
features.
“Madam,” said I civilly, “I hear you’re wed today to a gentleman
of standing.”
“What is that to you, sir?” she asked quickly.
“’Tis nothing, for sure,” said I, “but to a friend of mine that I
value deeply ’tis much.”
“You speak of Mr. Masters,” said she sharply, and with
discomposure. “Sure, if he be a gentleman, he will not trouble me
when he knows.”
“Anne!” said a voice from the top of the stairs, “Anne!”
’Twas her bridegroom calling. Well, she should go to him in what
mood she might when I had done with her.
“He will never know,” says I, “unless he hear it from yourself.”
“Anne!” said the voice above the stairs.
“He shall not—I will not,” she cried angrily. “I will not be
persecuted. ’Twas all a mistake.”
I whistled. Calypso emerged from the night, and behind Calypso
was the horse with its burden. An anxious look dawned in her face.
“I am insulted,” says she and paused quickly.
“Edward!” she called, and put a hand to her bosom.
“Anne, darling!” cried the voice, “where are you? Come, child, ’tis
late.”
The horse came to a stop before the door with the body on the
saddle, bound to the crupper.
“What is it?” she cried in alarm, and suddenly she shrieked,
recognizing what was there. “It is an omen—my wedding night!”
“Aye,” says I, “which be your bridegroom, he that calls or he that
is silent? Call on him and he hears not.”
Peal after peal went up now from her, and the house was awake
with alarm. I turned away, leaving her on the doorstep, and
mounted the mare.
As I cantered off into the night I cast a glance behind me, and a
group was gathered at the door, and in that group lay Mistress Anne
fallen in a swoon, with the sleeping figure on the horse before her.
The Judge and the Jack Tar
BY HENRY H. CORNISH
IT’S like this here, Your Honor, see?
As near as I can tell,
A gentleman hired my boat, and he
Was quite a proper swell.
He brought a lady down with him
To make a longish trip
And so we scrubbed her thoroughly—
Judge—The lady?
Tar—No! The ship
Well—cutting off my story short

To come to what befell
We started, but put back to port
Which much annoyed the swell.
She fell between two waterways
And got a nasty nip,
So we rigged her out with brand-new stays—
Judge—The lady?
Tar—No-o! The ship.
At last we put to sea again

And started for the west,
All spick and span without a stain
When all at once, I’m blest,
Her blooming timbers got misplaced,
Which quite upset the trip,
The water washed around her waist—
Judge—The lady’s?
Tar (nodding)—And the ship’s.
That’s all, I think, Your Honor, now,

I’ll state to you my claim.
Five hundred dollars, you’ll allow,
Won’t build her up the same.
Her rudder’s gone, her nose is broke,
Her flag I’ve had to dip
She’s lying now upon the mud—
Judge—The lady?
Tar—No-o-o-o! The ship.
Object, Matrimony
BY CAROLINE LOCKHART
W
ITH a turn of his red wrist, Porcupine Jim guided his horse in
and out among the badger holes which made riding dangerous
business on the Blackfoot Reservation. Perplexity and discontent
rested upon Porcupine’s not too lofty brow. Though he looked at the
badger holes and avoided them mechanically, he saw them not.
“Would you tank, would you tank,” he burst out finally in a voice
which rasped with irritation, “dat a girl like Belle Dashiel would
rudder have dat pigeon-toed, smart-Aleck breed dan me?”
Porcupine’s pinto cayuse threw back one ear and listened
attentively to the naïve conceit of his rider’s soliloquy.
“Look at me!” demanded Porcupine, changing the reins to his left
hand that he might make a more emphatic gesture with his right. “A
honest Swede, able to make fifteen dollars a day at my trade. Me as
has sheared sheep from Montany to the Argentine Republic, gittin’
bounced for dat lazy half-breed dat can’t hold a yob two mont’!”
Porcupine’s thoughts upon any subject were not varied, and he
burst forth at intervals with a reiteration of the same idea until he
came to the ridge where he could look down upon the house of
Dashiel, the squaw-man, who kept a sort of post office in a soapbox.
Porcupine had come twenty-five miles for his mail. Not that he
expected any, but to be gibed at by Belle Dashiel had the same
fascination for him that biting on a sore tooth has for a small boy.
Gradually the knowledge had come to his slow-working mind that
the half-breed girl’s interest in him rose solely from the fact that
John Laney was his partner in the assessment work which they were
doing in the mountains on a tenderfoot’s copper claim.
Laney’s father had been an Irish steamboat captain on Lake
Superior, his mother, a Chippewa squaw, and the cross had produced
an unusual type. The Indian blood which keeps a half-breed silent
and shy before strangers had no such effect upon Laney. His
prowess was his theme and his vanity was a byword on the
Reservation. He obtained his fashions from the catalogue of a
wholesale house in Chicago which furnishes the trusting pioneer with
the latest thing in oil drills or feather boas. It was common belief
that Laney’s high celluloid collar would some day cut his head off.
Laney’s waking hours were spent in planning schemes of
primitive crudeness whereby he might acquire affluence without
labor. In his dreams the tenderfoot tourist was generally the person
who was to remove him from penury.
“Hello, Porcupine!” called Belle Dashiel, coming to the door with a
pink bow pinned on a pompadour of amazing height.
“Hullo yourself!” replied Porcupine, elated at his ready wit and
the cordiality in her voice.
“How’s John?”
The smile faded from his face.
“Good ’nough,” he replied shortly.
“When’s he comin’ down?”
“Dunno. Any mail for me?”
“A letter and a paper.”
“Who could be writin’ to me?”
Porcupine looked surprised.
“Didn’t you expect nothin’?” Belle Dashiel’s eyes shone
mischievously.
“Yass, I tank, mebby.” A deeper red spread over the Swede’s
sunburned face.
He opened his letter and spelled it out laboriously, his chest
heaving with the effort.
“A man over in Chicago he tank I’m in turrible need of a pianny,”
he said in disgust, as he put the circular in the stove.
Porcupine lingered till the chill of the night air crept into the
sunshine of the September day. Then he put spurs to his patient
cayuse and hit the trail which led into the fastnesses of the Rockies.
The light was not quite gone when he happened to think of the
paper he had thrust in his coat-pocket. There might be news in it!
Bacon-Rind-Dick had told Two-Dog-Jack that there was a war over in
Jay-pan. Porcupine removed the wrapper and the words Wedding
Chimes stared him in the face.
As he read, he laid the reins on his horse’s neck and let the pinto
pick his own road. The matrimonial sheet opened up a vista of
romantic adventures and possibilities of which the Swede had never
dreamed. His imagination, which naturally was not a winged thing,
was fired until he saw himself leading to his shack up the North Fork
of the Belly River the fairest and richest lady in the land. All he had
to do was to send five dollars to Wedding Chimes and thus join their
matrimonial club. Upon the receipt of the five dollars, the editor
would send him the names and addresses of several ladies who
were all young, beautiful, wealthy and anxious to be married. He
could open a correspondence with one or all of them, and then
choose for his bride the lady whose letter appealed to him most.
Porcupine strained his eyes reading descriptions of lily-white
blondes and dashing brunettes. When he could see no longer, he
folded the precious paper and buttoned it inside his coat.
His cayuse was puffing up the steep mountain trail in the
darkness of the thick pines and spruces when Porcupine suddenly let
out a yell which startled the prowling lynx and made his pinto snort
with fright. It was a wild whoop of exultation. There had come to
Porcupine one of those rare revelations which have made men great.
He fairly glowed and tingled with the inspiration which had flashed
upon him as though someone had gone through his brain with a
lantern.
When he rode into camp, where Laney sat before the fire eating
bacon out of a frying-pan, Porcupine’s deep-set blue eyes were
shining like stars on a winter’s night.
“Yass, I got de greatest ting in de mail you ever see, I tank!”
Laney’s face expressed curiosity as the Swede sat down on a log
and turned his felt hat round and round upon his bullet-shaped head
—a trick he had when excited. With great deliberation and
impressiveness he produced the paper and handed it to Laney.
Laney set the frying-pan where his wolfhound could finish the bacon
and opened the paper.
“Young, beautiful, immensely rich; obj., mat.,” he read. Laney’s
eyes sparkled. He read for half an hour of successful weddings
brought about by the editorial Cupid. Porcupine at last roused him
from his absorption.
“Laney, I got a scheme, I tank. I’ll join up with one of dem clubs
and you carry out de corryspondance with one of dem ladies. You
are a better scholar den me and write a pooty goot letter. Den, if it
goes all right, I’ll go and see her and tell her I ain’t exactly de man
dat done de writin’, but I’m just as goot.
“’Tain’t no use for you to get into de club, because you are all the
same as promised to Belle Dashiel. Sure,” Porcupine went on, “Belle
ain’t rich nor beautiful like dem ladies in Weddin’ Chimes, but she’s a
goot little girl.
“Old Dashiel ain’t got more dan fifty head of beef cattle, and dey
say he got a lot of runts in de last Govermint issue, but a ting like
dat don’t cut no ice if you’re stuck on de girl.”
Laney moved uneasily and avoided Porcupine’s eyes.
“Now for me,” continued the Swede, “I can marry any millionaire
I want to.”
As soon as the mails could get it there, the editor of Wedding
Chimes received a neatly penciled and eloquent letter from one John
Laney, setting forth his especial needs and preferences, with
considerable stress laid upon the financial standing of the
matrimonial candidates.
The day the list was due Laney rode down for the mail. The
eagerness with which he took the letter from her hand did not
escape Belle Dashiel.
“Got a new girl, John?” she asked lightly, though she watched his
face with suspicious eyes.
“Perhaps,” replied Laney, and all her urging could not detain him.
By the light of the camp-fire Laney and Porcupine studied the list
of names and addresses sent from the office of the matrimonial
paper.
“This a-here one suits me,” said Laney. “‘Mayme Livingston, Oak
Grove, Iowa.’ It’s a toney-sounding name.”
“It’s me dat’s gittin’ married,” Porcupine suggested significantly.
“But Mayme’s all right, I tank. Go on ahead and write.”
So Laney, with the assistance of a sheet of ruled notepaper and a
lead pencil which he moistened frequently in order to shade
effectively, composed a letter which he and Porcupine regarded not
only as a model of cleverness but an achievement from a literary
point of view. The legal tone which gave it dignity was much
admired by Porcupine. The letter read:
Belly River, Mont.

Miss Mayme Livingston:
Dear Madam: Whereas I have paid up five dollars
and have the priveledge of writing to any lady on the
list sent from the aforesaid matrimonial paper, I, the
undersigned, have picked out you, Miss Mayme
Livingston party of the first part, obj. mat.
I am an American, five feet seven, and quite dark. I
am interested in copper mines and cattle. I can ride
anything that wears hair and last winter I killed two
silver-tips and a link. I am engaged somewhat in
trapping also. They say I am a tony dresser and I can
dance the Portland Fancy or any dance that I see once.
I play the juice-harp, mouth organ and accordian. I
have a kind disposition and would make a good
husband to any lady who had a little income of her
own.
Let me hear from you as soon as you get this and
tell me what you think of my writing.
Respy. Yrs.
John Laney.
In witness whereof that this letter is true I have
hereunto set my hand and fixed my seal.
Porcupine Jim X his mark.
The days which followed the mailing of the above composition

were the longest Laney and Porcupine had ever known. They
discussed Miss Livingston until they felt they knew her. Porcupine
thought she had black eyes, black hair, was inclined to stoutness,
but with a good “figger.”
The name of Livingston to Laney conjured up a vision of blonde
loveliness in red satin, slender, shapely, with several thousand dollars
in a handbag which she kept always with her.
Miss Livingston’s letter came with delightful promptness. There
was an angry glow in Belle Dashiel’s Indian eyes as she handed the
salmon-pink envelope to Laney.
“Who you writin’ to?” she demanded.
“Business,” replied Laney bruskly, and strode out of the house.
Porcupine, who had also come down, lingered a moment to tell
her she looked prettier each time that he saw her.
Miss Livingston’s letter read:
Mr. John Laney
deer sir. i take a few minutes to tell you how glad i was
to heer from you Away off in montana i have not got
Much Noos to rite but i will explain abot Myself i am a
suthoner and quite Dark to my Father was a rice
planter before the war which ruhined us i hav a good
Voice and sing in the Quire i danz most evry Danc goin
i have a Stiddy incom and make hansom presints to
annybody i Like if i met a perfect Genelman i wold
Marry him i cannot rite annymore Today bekaws i hay
Piz to make rite offen to
Miss Mayme Livingston

i think your Ritin is good i wish you wold send your
Fotegraf
Laney’s brow was clouded as he folded the letter. “She ain’t much
of a scholar,” he said. “You hardly ever see a scholar use little ‘i’s.’”
“What differunce does dat make when she’s got a stiddy
income?” replied Porcupine quickly. “And den what she said about
handsome presents. Sure, she’s a hairess, I tank.”
Laney brightened at these reminders, and immediately set about
composing another letter calculated to impress the wealthy, if
unlettered, Miss Livingston.
“Dear madam,” soon developed into “Dearest Mayme,” and “deer
sir” as speedily became “darlig John,” and, with each salmon-pink
envelope’s arrival, Laney’s coolness toward Belle Dashiel became
more marked.
“Porcupine,” said Laney, who had begun to show some reluctance
in reading the correspondence to his partner, “the lady is gettin’
oneasy to see me, and when we finish runnin’ that drift in the lead, I
think I’ll take a trip over to Iowa and see her.”
“But where do I come in, mebby?” demanded Porcupine.
“That’s what I’m goin’ for—to fix it up for you. Reely, Porcupine,”
and he looked critically at the rawboned Swede, whose hair stood up
like the quills on the animal from which he had received his
sobriquet, “it wouldn’t be right for you to break in on a lady without
givin’ her warning of what you was like.”
“I know I ain’t pooty,” replied Porcupine unperturbed, “but I can
make fifteen dollars a day at my trade.”
The tenderfoot’s assessment money went toward buying Laney a
wardrobe which almost any one of Laney’s relatives or friends would
have killed him in his sleep to possess.
A jeweler, advertising in Wedding Chimes, received an order for a
one-hundred-and-fifty-dollar scarfpin, to be paid for in instalments.
Porcupine, whose nature was singularly free from envy, could not
but feel a pang when he saw the large horseshoe of yellow
diamonds glittering in Laney’s red cravat.
Laney had read that no gentleman should think of venturing into
polite society without a “dress suit.” An order was sent for a seventy-
five-dollar suit of evening clothes to the Chicago firm from whom
they bought their mining tools. When the clothes arrived Laney
dressed himself in them one evening in their shack up the North
Fork of Belly River, and Porcupine’s face showed the admiration he
felt, as Laney strutted like a pheasant drumming on a log.
Laney, who numbered among his accomplishments the ability to
draw a rose or a horse so that almost anybody would know what it
was, gave an original touch to his costume by purchasing at the
Agency a brown broad-brimmed felt hat and painting a red rose
directly in front under the stiff brim.
When the drift was run and Laney’s wardrobe was complete, he
and the Swede set out across the Reservation to the railroad station.
“Pardner,” said Porcupine as he looked wistfully at the broadcloth
coat with satin revers and the tail sloped away like a grasshopper’s
wings, “dey ain’t a friend you got, but me, dat would trust you to do
their courtin’ for them togged out like dat—sure, dat’s so!”
There was a derisive glint in Laney’s small back eyes; he held the
slow-witted Swede in almost open contempt for his innocence.
Porcupine shook hands with him on the platform and wished him
good luck. “You’ll do your best for me, pard?” he asked anxiously.
“Trust me,” replied Laney gaily, intoxicated by the attention he
was receiving from the tourists in the Pullman car.
Porcupine stopped at Dashiel’s on his return. Belle Dashiel met
him at the door and her eyes were blazing. Without being able to
define the process of reasoning by which he arrived at the
conclusion, Porcupine felt that his brilliant plot stood an infinitely
better show of success that he did not find her in tears.
“Where’s he gone at?” She stamped her moccasined foot
imperiously.
“I wouldn’t like to say,” replied Porcupine in a voice which
denoted a wish to shield his partner and yet a noble, if unusual,
desire to tell the truth.
“Tell me!” she commanded, and she put her small hand on the
big Swede’s arm as though she would shake him.
“I tank,” answered Porcupine meekly; “I dunno, but I tank he’s
gone to get married.”
As Laney sat in the day coach in his evening clothes, his broad
hat tilted back from his coarse, swarthy face, a constant procession
filed through the aisle and every eye rested upon his smiling and

[FREE PDF sample] Data Mining for Design and Marketing 1st Edition Yukio Ohsawa ebooks

Uploaded by

Copyright:

Available Formats

[FREE PDF sample] Data Mining for Design and Marketing 1st Edition Yukio Ohsawa ebooks

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

[FREE PDF sample] Data Mining for Design and Marketing 1st Edition Yukio Ohsawa ebooks

Uploaded by

Copyright:

Available Formats

Get ebook downloads in full at ebookname.

Data Mining for Design and Marketing 1st Edition

Explore and download more ebook at https://ebookname.com

Visual Data Mining Techniques and Tools for Data

Data Warehousing Design and Advanced Engineering

Investigative Data Mining for Security and Criminal

A Higher Bid How to Transform Special Event Fundraising

The Java Virtual Machine Specification Java SE 8

Intelligent Systems for Automated Learning and

Bartimaeus Trilogy Book Three Ptolemy s Gate Jonathan

Netter s Pediatrics 1st Edition Todd Florin

© 2009 by Taylor & Francis Group, LLC

C7019.indb 1 12/18/08 11:32:57 AM

AIMS AND SCOPE

UNDERSTANDING COMPLEX DATASETS: Data Mining with Matrix Decompositions

© 2009 by Taylor & Francis Group, LLC

C7019.indb 2 12/18/08 11:32:57 AM

Data Mining for

© 2009 by Taylor & Francis Group, LLC

C7019.indb 3 12/18/08 11:32:57 AM

No claim to original U.S. Government works

International Standard Book Number-13: 978-1-4200-7019-4 (Hardcover)

Library of Congress Cataloging-in-Publication Data

Visit the Taylor & Francis Web site at

© 2009 by Taylor & Francis Group, LLC

C7019.indb 4 12/18/08 11:32:58 AM

Chapter 2. Reframing the Data-Mining Process................................................ 19

Chapter 3. The Use of Online Market Analysis Systems to Achieve

Chapter 4. Finding Hierarchical Patterns in Large POS Data Using

Chapter 5. A Method to Search ARX Model Orders and Its Application

Chapter 6. Data Mining for Improved Web Site Design

Chapter 7. Discourse Analysis and Creativity Support

Chapter 8. Data Crystallization with Human Interactions Applied

© 2009 by Taylor & Francis Group, LLC

C7019.indb 5 12/18/08 11:32:58 AM

Chapter 11. Analysis Framework for Knowledge Discovery Related to

Chapter 12. Association Bundle-Based Market Basket Analysis....................... 187

Chapter 13. Formal Concept Analysis with Attribute Priorities........................ 211

Chapter 14. Literature Categorization System for Automated Database

Chapter 15. A Data-Mining Framework for Designing Personalized

Chapter 16. An Adjacency Matrix Approach for Extracting

© 2009 by Taylor & Francis Group, LLC

C7019.indb 6 12/18/08 11:32:58 AM

Appendix B: A Maximal Cliques Enumeration Algorithm

© 2009 by Taylor & Francis Group, LLC

C7019.indb 7 12/18/08 11:32:58 AM

© 2009 by Taylor & Francis Group, LLC

C7019.indb 9 12/18/08 11:32:58 AM

Yukio Ohsawa and Katsutoshi Yada

© 2009 by Taylor & Francis Group, LLC

C7019.indb 10 12/18/08 11:32:59 AM

Katsutoshi Yada is a professor of management information systems in the Faculty

© 2009 by Taylor & Francis Group, LLC

C7019.indb 11 12/18/08 11:32:59 AM

David Bergner, Ph.D. Gary Gregory, B.Sc., M.B.A., Ph.D.

Daryl H. Hepting, Ph.D.

Kenta Fukata Daiji Kato

© 2009 by Taylor & Francis Group, LLC

C7019.indb 13 12/18/08 11:32:59 AM

Limin Lin, Ph.D. Takanobu Nakahara, M.A.

Appendix B: A Maximal Cliques Enumeration Algorithm