Chapter 1 Data Preprocessing

Data Types and Forms

Attribute-value data: Data types numeric, categorical (see the hierarchy for its relationship) static, dynamic (temporal) Other kinds of data distributed data text, Web, meta data images, audio/video

A1

A2

An

CS583, Bing Liu, UIC

Chapter 2: Data Preprocessing

Why preprocess the data? Data cleaning Data integration and transformation Data reduction Discretization Summary
3

CS583, Bing Liu, UIC

Why Data Preprocessing?

Data in the real world is dirty

incomplete: missing attribute values, lack of certain attributes of interest, or containing only aggregate data

e.g., occupation= e.g., Salary=-10 e.g., Age=42 Birthday=03/07/1997 e.g., Was rating 1,2,3, now rating A, B, C e.g., discrepancy between duplicate records

noisy: containing errors or outliers

inconsistent: containing discrepancies in codes or names

CS583, Bing Liu, UIC

Why Is Data Preprocessing Important?

No quality data, no quality mining results!

Quality decisions must be based on quality data

e.g., duplicate or missing data may cause incorrect or even misleading statistics.

Data preparation, cleaning, and transformation comprises the majority of the work in a data mining application (90%).

CS583, Bing Liu, UIC

Multi-Dimensional Measure of Data Quality

A well-accepted multi-dimensional view:

Accuracy Completeness Consistency Timeliness Believability Value added Interpretability Accessibility

CS583, Bing Liu, UIC

Major Tasks in Data Preprocessing

Data cleaning

Fill in missing values, smooth noisy data, identify or remove outliers and noisy data, and resolve inconsistencies Integration of multiple databases, or files Normalization and aggregation Obtains reduced representation in volume but produces the same or similar analytical results

Data integration

Data transformation

Data reduction

Data discretization (for numerical data)

CS583, Bing Liu, UIC

Chapter 2: Data Preprocessing

Why preprocess the data? Data cleaning Data integration and transformation Data reduction Discretization Summary
8

CS583, Bing Liu, UIC

Data Cleaning

Importance

Data cleaning is the number one problem in data warehousing

Data cleaning tasks

Fill in missing values

Resolve redundancy caused by data integration

CS583, Bing Liu, UIC

Missing Data

Data is not always available

E.g., many tuples have no recorded values for several attributes, such as customer income in sales data

Missing data may be due to

equipment malfunction inconsistent with other recorded data and thus deleted data not entered due to misunderstanding certain data may not be considered important at the time of entry not register history or changes of the data
10

CS583, Bing Liu, UIC

How to Handle Missing Data?

Ignore the tuple Fill in missing values manually: tedious + infeasible? Fill in it automatically with

a global constant : e.g., unknown, a new class?! the attribute mean the most probable value: inference-based such as Bayesian formula, decision tree, or EM algorithm

CS583, Bing Liu, UIC

11

Noisy Data

Noise: random error or variance in a measured variable. Incorrect attribute values may due to

faulty data collection instruments data entry problems data transmission problems etc duplicate records, incomplete data, inconsistent data
12

Other data problems which requires data cleaning

CS583, Bing Liu, UIC

How to Handle Noisy Data?

Binning method:

first sort data and partition into (equi-depth) bins then one can smooth by bin means, smooth by bin median, smooth by bin boundaries, etc. detect and remove outliers detect suspicious values and check by human (e.g., deal with possible outliers)

Clustering

Combined computer and human inspection

CS583, Bing Liu, UIC

13

Binning Methods for Data Smoothing

Sorted data for price (in dollars): 4, 8, 9, 15, 21, 21, 24, 25, 26, 28, 29, 34 Partition into (equi-depth) bins:

Bin 1: 4, 8, 9, 15 Bin 2: 21, 21, 24, 25 Bin 3: 26, 28, 29, 34 Bin 1: 9, 9, 9, 9 Bin 2: 23, 23, 23, 23 Bin 3: 29, 29, 29, 29 Bin 1: 4, 4, 4, 15 Bin 2: 21, 21, 25, 25 Bin 3: 26, 26, 26, 34

Smoothing by bin means:

Smoothing by bin boundaries:

CS583, Bing Liu, UIC

14

Outlier Removal

Data points inconsistent with the majority of data Different outliers

Valid: CEOs salary, Noisy: Ones age = 200, widely deviated points Clustering Curve-fitting Hypothesis-testing with a given model

Removal methods

CS583, Bing Liu, UIC

15

Chapter 2: Data Preprocessing

Why preprocess the data? Data cleaning Data integration and transformation Data reduction Discretization Summary
16

CS583, Bing Liu, UIC

Data Integration

Data integration:

combines data from multiple sources integrate metadata from different sources Entity identification problem: identify real world entities from multiple data sources, e.g., A.cust-id B.cust-# for the same real world entity, attribute values from different sources are different, e.g., different scales, metric vs. British units

Schema integration

Detecting and resolving data value conflicts

Removing duplicates and redundant data

CS583, Bing Liu, UIC

Data Transformation

Smoothing: remove noise from data

Attribute/feature construction

New attributes constructed from the given ones

Aggregation: summarization

Generalization: concept hierarchy climbing

CS583, Bing Liu, UIC

18

Data Transformation: Normalization

min-max normalization

v minA v' (new _ maxA new _ minA) new _ minA maxA minA z-score normalization

v meanA v' stand _ devA normalization by decimal scaling

CS583, Bing Liu, UIC

19

Chapter 2: Data Preprocessing

Why preprocess the data?

CS583, Bing Liu, UIC

Data Reduction Strategies

Data is too big to work with Data reduction

Obtain a reduced representation of the data set that is much smaller in volume but yet produce the same (or almost the same) analytical results Dimensionality reduction remove unimportant attributes Aggregation and clustering Sampling

Data reduction strategies

CS583, Bing Liu, UIC

21

Dimensionality Reduction

Feature selection (i.e., attribute subset selection):

Select a minimum set of attributes (features) that is sufficient for the data mining task. step-wise forward selection step-wise backward elimination combining forward selection and backward elimination etc

Heuristic methods (due to exponential # of choices):

CS583, Bing Liu, UIC

22

Histograms
40

35 A popular data reduction technique 30 Divide data into buckets 25 and store average (sum) for each bucket 20 15 10 5 0

CS583, Bing Liu, UIC

10000

30000

50000

70000

90000 23

Clustering

Partition data set into clusters, and one can store cluster representation only

Can be very effective if data is clustered but not if

data is smeared

There are many choices of clustering definitions and clustering algorithms. We will discuss them later.

CS583, Bing Liu, UIC

24

Sampling

Choose a representative subset of the data

Simple random sampling may have poor performance in the presence of skew. Stratified sampling: Approximate the percentage of each class (or subpopulation of interest) in the overall database Used in conjunction with skewed data

Develop adaptive sampling methods

CS583, Bing Liu, UIC

25

Sampling
Raw Data
Cluster/Stratified Sample

CS583, Bing Liu, UIC

26

Chapter 2: Data Preprocessing

Why preprocess the data?

CS583, Bing Liu, UIC

Discretization

Three types of attributes:

Nominal values from an unordered set Ordinal values from an ordered set Continuous real numbers divide the range of a continuous attribute into intervals because some data mining algorithms only accept categorical attributes. Binning methods equal-width, equal-frequency Entropy-based methods
28

Discretization:

Some techniques:

CS583, Bing Liu, UIC

Discretization and Concept Hierarchy

Discretization

reduce the number of values for a given continuous attribute by dividing the range of the attribute into intervals. Interval labels can then be used to replace actual data values reduce the data by collecting and replacing low level concepts (such as numeric values for the attribute age) by higher level concepts (such as young, middle-aged, or senior)

Concept hierarchies

CS583, Bing Liu, UIC

29

Binning

Attribute values (for one attribute e.g., age):

0, 4, 12, 16, 16, 18, 24, 26, 28 Bin 1: 0, 4 [-,10) bin Bin 2: 12, 16, 16, 18 [10,20) bin Bin 3: 24, 26, 28 [20,+) bin denote negative infinity, + positive infinity Bin 1: 0, 4, 12 Bin 2: 16, 16, 18 Bin 3: 24, 26, 28 [-, 14) bin [14, 21) bin [21,+] bin
30

Equi-width binning for bin width of e.g., 10:

Equi-frequency binning for bin density of e.g., 3:

CS583, Bing Liu, UIC

Entropy-based (1)

Given attribute-value/class pairs:

(0,P), (4,P), (12,P), (16,N), (16,N), (18,P), (24,N), (26,N), (28,N) Intuitively, find best split so that the bins are as pure as possible Formally characterized by maximal information gain.

Entropy-based binning via binarization:

Let S denote the above 9 pairs, p=4/9 be fraction of P pairs, and n=5/9 be fraction of N pairs. Entropy(S) = - p log p - n log n.

Smaller entropy set is relatively pure; smallest is 0. Large entropy set is mixed. Largest is 1.

CS583, Bing Liu, UIC

31

Entropy-based (2)

Let v be a possible split. Then S is divided into two sets: S1: value <= v and S2: value > v Information of the split: I(S1,S2) = (|S1|/|S|) Entropy(S1)+ (|S2|/|S|) Entropy(S2) Information gain of the split: Gain(v,S) = Entropy(S) I(S1,S2) Goal: split with maximal information gain. Possible splits: mid points b/w any two consecutive values. For v=14, I(S1,S2) = 0 + 6/9*Entropy(S2) = 6/9 * 0.65 = 0.433 Gain(14,S) = Entropy(S) - 0.433 maximum Gain means minimum I. The best split is found after examining all possible splits.

CS583, Bing Liu, UIC

32

Summary

Data preparation is a big issue for data mining

Data cleaning and data integration

Data reduction and feature selection

Many methods have been proposed but still an active area of research

CS583, Bing Liu, UIC

33