Oracle Data Mining

Oracle Data Mining (ODM) is an option of Oracle Database Enterprise Edition. It contains several data
mining and data analysis algorithms for classification, prediction, regression, associations, feature selection,
anomaly detection, feature extraction, and specialized analytics. It provides means for the creation,
management and operational deployment of data mining models inside the database environment.

Overview Oracle Data Mining

Developer(s) Oracle Corporation
Oracle Corporation has implemented a variety of data mining
Stable release 11gR2 /
algorithms inside its Oracle Database relational database product.
These implementations integrate directly with the Oracle database September, 2009
kernel and operate natively on data stored in the relational database Type data mining and
tables. This eliminates the need for extraction or transfer of data analytics
into standalone mining/analytic servers. The relational database
License proprietary
platform is leveraged to securely manage models and to efficiently
execute SQL queries on large volumes of data. The system is Website Oracle Data Mining
organized around a few generic operations providing a general (http://www.oracle.
unified interface for data-mining functions. These operations com/technetwork/d
include functions to create, apply, test, and manipulate data-mining atabase/options/ad
models. Models are created and stored as database objects, and vanced-analytics/o
their management is done within the database - similar to tables, dm/index.html)
views, indexes and other database objects.

In data mining, the process of using a model to derive predictions or descriptions of behavior that is yet to
occur is called "scoring". In traditional analytic workbenches, a model built in the analytic engine has to be
deployed in a mission-critical system to score new data, or the data is moved from relational tables into the
analytical workbench - most workbenches offer proprietary scoring interfaces. ODM simplifies model
deployment by offering Oracle SQL functions to score data stored right in the database. This way, the
user/application-developer can leverage the full power of Oracle SQL - in terms of the ability to pipeline
and manipulate the results over several levels, and in terms of parallelizing and partitioning data access for
performance.

Models can be created and managed by one of several means. Oracle Data Miner provides a graphical user
interface that steps the user through the process of creating, testing, and applying models (e.g. along the
lines of the CRISP-DM methodology). Application- and tools-developers can embed predictive and
descriptive mining capabilities using PL/SQL or Java APIs. Business analysts can quickly experiment with,
or demonstrate the power of, predictive analytics using Oracle Spreadsheet Add-In for Predictive Analytics,
a dedicated Microsoft Excel adaptor interface. ODM offers a choice of well-known machine learning
approaches such as Decision Trees, Naive Bayes, Support vector machines, Generalized linear model
(GLM) for predictive mining, Association rules, K-means and Orthogonal Partitioning[1][2] Clustering, and
Non-negative matrix factorization for descriptive mining. A minimum description length based technique to
grade the relative importance of input mining attributes for a given problem is also provided. Most Oracle
Data Mining functions also allow text mining by accepting text (unstructured data) attributes as input. Users
do not need to configure text-mining options - the Database_options database option handles this behind
the scenes.

History
Oracle Data Mining was first introduced in 2002 and its releases are named according to the corresponding
Oracle database release:

Oracle Data Mining 9iR2 (9.2.0.1.0 - May 2002)

Oracle Data Mining 10gR1 (10.1.0.2.0 - February 2004)
Oracle Data Mining 10gR2 (10.2.0.1.0 - July 2005)
Oracle Data Mining 11gR1 (11.1 - September 2007)
Oracle Data Mining 11gR2 (11.2 - September 2009)

Oracle Data Mining is a logical successor of the Darwin data mining toolset developed by Thinking
Machines Corporation in the mid-1990s and later distributed by Oracle after its acquisition of Thinking
Machines in 1999. However, the product itself is a complete redesign and rewrite from ground-up - while
Darwin was a classic GUI-based analytical workbench, ODM offers a data mining
development/deployment platform integrated into the Oracle database, along with the Oracle Data Miner
GUI.

The Oracle Data Miner 11gR2 New Workflow GUI was previewed at Oracle Open World 2009. An
updated Oracle Data Miner GUI was released in 2012. It is free, and is available as an extension to Oracle
SQL Developer 3.1 .

Functionality
As of release 11gR1 Oracle Data Mining contains the following data mining functions:

Data transformation and model analysis:

Data sampling, binning, discretization, and other data transformations.
Model exploration, evaluation and analysis.
Feature selection (Attribute Importance).
Minimum description length (MDL).
Classification.
Naive Bayes (NB).
Generalized linear model (GLM) for Logistic regression.
Support Vector Machine (SVM).
Decision Trees (DT).
Anomaly detection.
One-class Support Vector Machine (SVM).
Regression
Support Vector Machine (SVM).
Generalized linear model (GLM) for Multiple regression
Clustering:
 Enhanced k-means (EKM).
Orthogonal Partitioning Clustering (O-Cluster).[1][2]
Association rule learning:
Itemsets and association rules (AM).
Feature extraction.
Non-negative matrix factorization (NMF).
Text and spatial mining:
Combined text and non-text columns of input data.
Spatial/GIS data.

Input sources and data preparation

Most Oracle Data Mining functions accept as input one relational table or view. Flat data can be combined
with transactional data through the use of nested columns, enabling mining of data involving one-to-many
relationships (e.g. a star schema). The full functionality of SQL can be used when preparing data for data
mining, including dates and spatial data.

Oracle Data Mining distinguishes numerical, categorical, and unstructured (text) attributes. The product
also provides utilities for data preparation steps prior to model building such as outlier treatment,
discretization, normalization and binning (sorting in general speak)

Graphical user interface: Oracle Data Miner

Users can access Oracle Data Mining through Oracle Data Miner, a GUI client application that provides
access to the data mining functions and structured templates (called Mining Activities) that automatically
prescribe the order of operations, perform required data transformations, and set model parameters. The
user interface also allows the automated generation of Java and/or SQL code associated with the data-
mining activities. The Java Code Generator is an extension to Oracle JDeveloper. An independent interface
also exists: the Spreadsheet Add-In for Predictive Analytics which enables access to the Oracle Data
Mining Predictive Analytics PL/SQL package from Microsoft Excel.

From version 11.2 of the Oracle database, Oracle Data Miner integrates with Oracle SQL Developer.[3]

PL/SQL and Java interfaces

Oracle Data Mining provides a native PL/SQL package (DBMS_DATA_MINING) to create, destroy,
describe, apply, test, export and import models. The code below illustrates a typical call to build a
classification model:

BEGIN
DBMS_DATA_MINING.CREATE_MODEL (
model_name => 'credit_risk_model',
function => DBMS_DATA_MINING.classification,
data_table_name => 'credit_card_data',
case_id_column_name => 'customer_id',
target_column_name => 'credit_risk',
settings_table_name => 'credit_risk_model_settings');
END;
where 'credit_risk_model' is the model name, built for the express purpose of classifying future customers'
'credit_risk', based on training data provided in the table 'credit_card_data', each case distinguished by a
unique 'customer_id', with the rest of the model parameters specified through the table
'credit_risk_model_settings'.

Oracle Data Mining also supports a Java API consistent with the Java Data Mining (JDM) standard for data
mining (JSR-73) for enabling integration with web and Java EE applications and to facilitate portability
across platforms.

SQL scoring functions

As of release 10gR2, Oracle Data Mining contains built-in SQL functions for scoring data mining models.
These single-row functions support classification, regression, anomaly detection, clustering, and feature
extraction. The code below illustrates a typical usage of a classification model:

SELECT customer_name
FROM credit_card_data
WHERE PREDICTION (credit_risk_model USING *) = 'LOW' AND customer_value = 'HIGH';

PMML
In Release 11gR2 (11.2.0.2), ODM supports the import of externally created PMML for some of the data
mining models. PMML is an XML-based standard for representing data mining models.

Predictive analytics Microsoft Excel add-in

The PL/SQL package DBMS_PREDICTIVE_ANALYTICS automates the data mining process including
data preprocessing, model building and evaluation, and scoring of new data. The PREDICT operation is
used for predicting target values classification or regression while EXPLAIN ranks attributes in order of
influence in explaining a target column feature selection. The new 11g feature PROFILE finds customer
segments and their profiles, given a target attribute. These operations can be used as part of an operational
pipeline providing actionable results or displayed for interpretation by end users.

References and further reading

T. H. Davenport, Competing on Analytics (https://web.archive.org/web/20110605215537/htt
p://www.lbl.gov/BLI/BLI_Library/assets/articles/OM/OM_PSDM_Competing_Analytics.pdf),
Harvard Business Review, January 2006.
I. Ben-Gal,Outlier detection (http://www.eng.tau.ac.il/~bengal/outlier.pdf), In: Maimon O. and
Rockach L. (Eds.) Data Mining and Knowledge Discovery Handbook: A Complete Guide for
Practitioners and Researchers," Kluwer Academic Publishers, 2005, ISBN 0-387-24435-2.
M. M. Campos, P. J. Stengard, and B. L. Milenova, Data-centric Automated Data Mining. In
proceedings of the Fourth International Conference on Machine Learning and Applications
2005, 15–17 December 2005. pp8, ISBN 0-7695-2495-8
M. F. Hornick, Erik Marcade, and Sunil Venkayala. Java Data Mining: Strategy, Standard,
and Practice. Morgan-Kaufmann, 2006, ISBN 0-12-370452-9.
B. L. Milenova, J. S. Yarmus, and M. M. Campos. SVM in Oracle database 10g: removing the
barriers to widespread adoption of support vector machines. In Proceedings of the 31st
 international Conference on Very Large Data Bases (Trondheim, Norway, August 30 -
September 2, 2005). pp1152–1163, ISBN 1-59593-154-6.
B. L. Milenova and M. M. Campos. O-Cluster: scalable clustering of large high dimensional
data sets. In proceedings of the 2002 IEEE International Conference on Data Mining: ICDM
2002. pp290–297, ISBN 0-7695-1754-4.
P. Tamayo, C. Berger, M. M. Campos, J. S. Yarmus, B. L.Milenova, A. Mozes, M. Taft, M.
Hornick, R. Krishnan, S.Thomas, M. Kelly, D. Mukhin, R. Haberstroh, S. Stephens and J.
Myczkowski. Oracle Data Mining - Data Mining in the Database Environment. In Part VII of
Data Mining and Knowledge Discovery Handbook, Maimon, O.; Rokach, L. (Eds.) 2005,
p315-1329, ISBN 0-387-24435-2.
Brendan Tierney, Predictive Analytics using Oracle Data Miner: for the data scientist, oracle
analyst, oracle developer & DBA, Oracle Press, McGraw Hill, Spring 2014.

See also
Oracle LogMiner - in contrast to generic data mining, targets the extraction of information
from the internal logs of an Oracle database

References
1. US patent 7174344 (https://worldwide.espacenet.com/textdoc?DB=EPODOC&IDX=US7174
344), Campos, Marcos M. & Milenova, Boriana L., "Orthogonal partitioning clustering",
issued 2007-02-06, assigned to Oracle International Corporation
2. Boriana L. Milenova and Marcos M. Campos (2002); O-Cluster: Scalable Clustering of Large
High Dimensional Data Sets (http://dl.acm.org/citation.cfm?id=844729), ICDM '02
Proceedings of the 2002 IEEE International Conference on Data Mining, pages 290-297,
ISBN 0-7695-1754-4.
3. "Oracle Data Miner" (http://www.oracle.com/technetwork/database/options/advanced-analyti
cs/odm/dataminerworkflow-168677.html?msgid=3-10214982236). Oracle technology
Network. Oracle Corporation. 2014. Retrieved 2014-07-17. "The Oracle Data Miner is an
Oracle SQL Developer extension that enables data analysts to work directly with data inside
the database, explore the data graphically, build and evaluate multiple data mining models,
apply Oracle Data Mining models to new data and deploy Oracle Data Mining's predictions
and insights throughout the enterprise. [...] Oracle Data Miner is comprised of three
components: Oracle Database 12c or Oracle Database 11g Release 2 SQL Developer
(client) which bundles the Oracle Data Miner work flow GUI Data Miner Repository -
installed in the Oracle Database"

External links
Oracle Data Mining at Oracle Technology Network (http://www.oracle.com/technology/produ
cts/bi/odm/index.html).
Oracle Data Mining Blog (http://blogs.oracle.com/datamining/).
Oracle Database 11g at Oracle Technology Network (http://www.oracle.com/technology/prod
ucts/database/oracle11g/index.html).
Oracle Data Mining and Analytics Blog (http://oracledmt.blogspot.com/).
Oracle Wiki for Data Mining (http://wiki.oracle.com/page/Oracle+Data+Mining).
Oracle Data Mining RSS Feed (http://www.oracle.com/technology/syndication/rss_otn_odm.
xml).
 Oracle Data Mining at Oracle Technology Network (http://www.oracle.com/technology/produ
cts/bi/odm/index.html).
Oracle Data Mining related blog by Brendan Tierney (Oracle ACE Director) (http://www.oralyt
ics.com).
Oracle Data Mining Examples (on Panoply Technology) (https://panoplytechen.wordpress.c
om/tag/odm/).

Retrieved from "https://en.wikipedia.org/w/index.php?title=Oracle_Data_Mining&oldid=1163702534"