Data Mining:

Concepts and Techniques

Chapter 3: Data Preprocessing

October 28, 2021 Data Mining: Concepts and Techniques 1

 Chapter 3: Data Preprocessing

 Why preprocess the data?

 Data cleaning
 Data integration and transformation
 Data reduction
 Discretization and concept hierarchy generation
 Summary

October 28, 2021 Data Mining: Concepts and Techniques 2

 Why Data Preprocessing?

 Data in the real world is dirty

 incomplete: lacking attribute values, lacking certain

attributes of interest, or containing only aggregate

data – e.g. Sales for advertising, cust don’t give info
 noisy: containing errors or outliers – machine faulty

 e.g., Salary=“-10”
 inconsistent: containing discrepancies in codes or
names – 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
 No quality data, no quality mining results!
 Quality decisions must be based on quality data

October 28, 2021 Data Mining: Concepts and Techniques 3

 Multi-Dimensional Measure of Data Quality

 A well-accepted multidimensional view:

 Accuracy

 Completeness

 Consistency

 Timeliness

 Believability

 Value added

 Interpretability

 Accessibility

 Broad categories:
 intrinsic, contextual, representational, and

accessibility.
October 28, 2021 Data Mining: Concepts and Techniques 4
 Major Tasks in Data Preprocessing

 Data cleaning
 Fill in missing values, smooth noisy data, identify or remove
outliers, and resolve inconsistencies
 Data integration
 Integration of multiple databases, data cubes, or files
 Data transformation
 Normalization and aggregation
 Data reduction
 Obtains reduced representation in volume but produces the same
or similar analytical results
 Data discretization
 Part of data reduction but with particular importance, especially for
numerical data

October 28, 2021 Data Mining: Concepts and Techniques 5

 Forms of data preprocessing

October 28, 2021 Data Mining: Concepts and Techniques 6

 Chapter 3: Data Preprocessing

 Why preprocess the data?

 Data cleaning
 Data integration and transformation
 Data reduction
 Discretization and concept hierarchy generation
 Summary

October 28, 2021 Data Mining: Concepts and Techniques 7

 Data Cleaning

 Data cleaning tasks

 Fill in missing values
 Identify outliers and smooth out noisy data
 Correct inconsistent data

October 28, 2021 Data Mining: Concepts and Techniques 8

 Missing Data
 Data is not always available
 E.g., many tuples have no recorded value 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
 Missing data may need to be inferred.

October 28, 2021 Data Mining: Concepts and Techniques 9

 How to Handle Missing Data?
 Ignore the tuple: usually done when class label is missing (assuming
the tasks in classification—not effective when the percentage of
missing values per attribute varies considerably.
 Fill in the missing value manually: tedious + infeasible
 Use a global constant to fill in the missing value: e.g., “unknown”, ~
(∞ value)
 Use the attribute mean to fill missing value – eg. Use avg income
 Use the attribute mean for all samples belonging to the same class to
fill in the missing value: smarter
 Use the most probable value to fill in the missing value: inference-
based such as Bayesian formula or decision tree

October 28, 2021 Data Mining: Concepts and Techniques 10

 How to Handle Missing Data?

Age Income Team Gender

23 24,200 Red Sox M
39 ? Yankees F
45 45,390 ? F

Fill missing values using aggregate functions (e.g., average) or

probabilistic estimates on global value distribution
E.g., put the average income here, or put the most probable income
based on the fact that the person is 39 years old
E.g., put the most frequent team here
 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

 technology limitation

 inconsistency in naming convention

 Other data problems which requires data cleaning

 duplicate records

 incomplete data

 inconsistent data

October 28, 2021 Data Mining: Concepts and Techniques 12

 How to Handle Noisy Data?
Smoothing techniques

 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.

 Clustering
 detect and remove outliers

 Combined computer and human inspection

 computer detects suspicious values, which are then

checked by humans
 Regression
 smooth by fitting the data into regression functions
 Simple Discretization Methods: Binning

 Equal-width (distance) partitioning:

 It divides the range into N intervals of equal size: uniform grid
 if A and B are the lowest and highest values of the attribute, the
width of intervals will be: W = (B-A)/N.
 The most straightforward
 But outliers may dominate presentation
 Skewed data is not handled well.
 Equal-depth (frequency) partitioning:
 It divides the range into N intervals, each containing
approximately same number of samples
 Good data scaling – good handing of skewed data
 Simple Discretization Methods: Binning

number
of values
Example: customer ages

Equi-width
binning: 0-10 10-20 20-30 30-40 40-50 50-60 60-70 70-80

Equi-depth
binning: 0-22 22-31 62-80
38-44 48-55
32-38 44-48 55-62
Smoothing using Binning Methods

* Sorted data for price (in dollars): 4, 8, 9, 15, 21, 21, 24, 25, 26, 28,
29, 34
* Partition into (equal) bins:
- Bin 1: 4, 8, 9, 15
- Bin 2: 21, 21, 24, 25
- Bin 3: 26, 28, 29, 34
* Smoothing by bin means:
- Bin 1: 9, 9, 9, 9
- Bin 2: 23, 23, 23, 23
- Bin 3: 29, 29, 29, 29
* Smoothing by bin boundaries: [4,15],[21,25],[26,34]
- Bin 1: 4, 4, 4, 15
- Bin 2: 21, 21, 25, 25
- Bin 3: 26, 26, 26, 34
 •Sorted data for price (in dollars):
•13,15,16, 16,19,20,22,25,25,25,33,33,35,35,52,70

•  Partition into (equal) bins:

      - Bin 1: 13,15,16
Binning       - Bin 2: 16,19,20
      - Bin 3: 22,25,25
Method       - Bin 4: 25,25,33
      - Bin 5: 33,35,35
      - Bin 6: 52,70

Smoothing [bin means:] Smoothing [bin medians:] Smoothing [bin boundaries:]

      - Bin 1: 14,14,14       - Bin 1: 15,15,15     - Bin 1: 13,16,16
      - Bin 2: 18,18,18       - Bin 2: 19,19,19     - Bin 2: 16,20,20
      - Bin 4: 24,24,24       - Bin 3: 25,25,25     - Bin 3: 22,25,25
      - Bin 3: 27,27,27       - Bin 4: 25,25,25     - Bin 4: 25,25,33
      - Bin 4: 34,34,34       - Bin 5: 35,35,35     - Bin 5: 33,35,35
      - Bin 5: 61,61       - Bin 6: 61,61     - Bin 6: 52,70

October 28, 2021 Data Mining: Concepts and Techniques 17

 Cluster Analysis

salary

cluster

outlier

age
 Regression
y (salary)

Example of linear regression

Y1 y=x+1

X1 x (age)
 Smoothing Noisy Data
 The purpose of data smoothing is to eliminate noise

Bin1: 4, 8, 15
Binning Bin2: 21, 21, 24
Bin3: 25, 28, 34
Clustering
means boundaries
Regression Bin1: 9, 9, 9 Bin1: 4, 4, 15
Bin2: 22, 22, Bin2: 21, 21, 24
22 Bin3: 25, 25, 34
Bin3: 29, 29,
29

20
 Chapter 3: Data Preprocessing

 Why preprocess the data?

 Data cleaning
 Data integration and transformation
 Data reduction
 Discretization and concept hierarchy generation
 Summary

October 28, 2021 Data Mining: Concepts and Techniques 21

 Data Integration

 Data integration:
 Data analysis may require combination of data from multiple
sources into a coherent data store
 Schema integration
 integrate metadata from different sources
 metadata: data about the data (i.e., data descriptors)
 Entity identification problem: identify real world entities from
multiple data sources, e.g., A.cust-id  B.cust-#
 Detecting and resolving data value conflicts
 for the same real world entity, attribute values from different
sources are different (e.g., J.D.Smith and Jonh Smith may refer to
the same person)
 possible reasons: different representations, different scales, e.g.,
(inches vs. cm)
 Handling Redundant Data
in Data Integration
 Redundant data occur often when integration of multiple
databases
 The same attribute may have different names in
different databases
 One attribute may be a “derived” attribute in another
table, e.g. age is derived from DOB
 Redundant data may be able to be detected by
correlational analysis
 Careful integration of the data from multiple sources may
help to reduce redundancies and inconsistencies and
improve mining speed and quality
October 28, 2021 Data Mining: Concepts and Techniques 23
 Data Transformation

 Smoothing: remove noise from data

 Aggregation: summarization, data cube construction
 Generalization: concept hierarchy climbing
 Normalization: scaled to fall within a small, specified
range
 min-max normalization
 z-score normalization
 normalization by decimal scaling
 Attribute/feature construction
 New attributes constructed from the given ones
 Data Transformation: Normalization

 min-max normalization
v  minA
v'  (new _ maxA  new _ minA)  new _ minA
maxA  minA
 e.g. convert age=30 to range 0-1, when min=10,max=80.
new_age=(30-10)/(80-10)=2/7
 z-score normalization v  meanA
v' 
stand_devA
 normalization by decimal scaling
v
v'  j Where j is the smallest integer such that Max(| v ' |)<1
10
 Chapter 3: Data Preprocessing

 Why preprocess the data?

 Data cleaning
 Data integration and transformation
 Data reduction
 Discretization and concept hierarchy generation
 Summary

October 28, 2021 Data Mining: Concepts and Techniques 26

 Data Reduction Strategies
 Warehouse may store terabytes of data: Complex data
analysis/mining may take a very long time to run on the
complete data set
 Data reduction
 Obtains a reduced representation of the data set that is

much smaller in volume but yet produces the same (or

almost the same) analytical results
 Data reduction strategies
 Data cube aggregation

 Dimensionality reduction

 Numerosity reduction

 Discretization and concept hierarchy generation

October 28, 2021 Data Mining: Concepts and Techniques 27

 Data Cube Aggregation

 Data Cube stores multidimensional aggregated

information.
 The cube created at lowest level of a abstraction - base
cuboid - the aggregated data for an individual entity of
interest
 The cube at the highest level of abstraction – apex cuboid
 Queries regarding aggregated information should be
answered using data cube, when possible

October 28, 2021 Data Mining: Concepts and Techniques 28

 Data Cube Aggregation

Sales data for given branch of A Data-Cube for sales at AllElectronics

AllElectronics for the years (Apex cuboid)
2002 to 2004

October 28, 2021 Data Mining: Concepts and Techniques 29

 Dimensionality Reduction
 Feature selection (i.e., Attribute Subset Selection):
 Select only the necessary attributes.
 The goal is to find a minimum set of attributes such that the resulting probability
distribution of data classes is as close as possible to the original distribution obtained
using all attributes.
 Eg. Classify customer as to whether or not they buy new popular CD on sale,
at that time, customer phone num. is irrelevant but age and music_test is revelant
The best(and worst) attributes are determined using test of statistical significance.

 Use heuristics: select local ‘best’ (or most pertinent) attribute, e.g.,
using information gain, etc. Eg. Initial Set -{A1, A2, A3, A4, A5, A6}
 step-wise forward selection {}{A1}{A1, A4}{A1, A4, A6}
 step-wise backward elimination {A1, A2, A3, A4, A5, A6}{A1, A3, A4, A5, A6}
{A1, A4, A5, A6} – Reduced set - > {A1, A4, A6}
 combining forward selection and backward elimination
 decision-tree induction
October 28, 2021 Data Mining: Concepts and Techniques 30
 Example of Decision Tree Induction

Decision Tree Induction constructs flow chart like structure

Initial attribute set: {A1, A2, A3, A4, A5, A6}

A4 ?

A1? A6?

Class 1 Class 2 Class 1 Class 2

> Reduced attribute set: {A1, A4, A6}

October 28, 2021 Data Mining: Concepts and Techniques 31

 Data Compression
 Data encoding or transformations are applied so as to obtain
a reduced or compressed representation of original data -
either lossy or lossless
 String compression
 There are extensive theories and well-tuned algorithms

 Typically lossless

 But only limited manipulation is possible without

expansion
 Audio/video compression
 Typically lossy compression, with progressive refinement

 Sometimes small fragments of signal can be

reconstructed without reconstructing the whole

October 28, 2021 Data Mining: Concepts and Techniques 32
 Data Compression

Original Data Compressed

Data
lossless

ss y
lo
Original Data
Approximated

October 28, 2021 Data Mining: Concepts and Techniques 33

 Compression

We could use a lossy compression technique to remove

attributes that are considered not important enough to
keep. Consider a 100% jpeg and a 85% jpeg... they
look very similar, but some unnecessary information is
lost... the sort of thing we want to do with attributes.

Techniques:
DWT: Discrete Wavelet Transform
DFT: Discrete Fourier Transform
PCA: Principal Component Analysis

October 28, 2021 Data Mining: Concepts and Techniques 34

 Discrete Wavelet Transforms
 DWT transforms a vector of attribute values into a different vector
of 'wavelet coefficients', of the same length as the original.

 This transformed vector can be truncated at a certain threshold. Any

values below the threshold are set to 0.

 The remaining data is then an approximation of the original, in the

transformed space.

 We can reverse the transformation to return to the original

attributes, minus the ones lost in the truncation.

 There are many different DWTs, in families (eg Haar and

Daubechies)

October 28, 2021 Data Mining: Concepts and Techniques 35

 Wavelet Transforms
 Method:
 Length, L, must be an integer power of 2 (padding with 0s, when
necessary)
 Each transform has 2 functions: smoothing, difference
 Applies to pairs of data, resulting in two set of data of length L/2
 Applies two functions recursively, until reaches the desired length

Haar2 Daubechie4

October 28, 2021 Data Mining: Concepts and Techniques 36

 Principal Component Analysis
Not a signal processing technique!
Idea: Just because the dataset has various dimensional
axes, doesn't mean those are the best axes to use.
Find the best axes in order and drop the least important
ones

Dataset on regular axes Dataset on revised axes

October 28, 2021 Data Mining: Concepts and Techniques 37
 Principal Component Analysis
 Do not reduce attribute as in the “attribute subset selection”
 Suppose that the data to be reduced consist of tuples or
data vectors described by n attributes or dimensions.
 Principal components analysis, or PCA (also called the
Karhunen-Loeve, or K-L, method), searches for k n-
dimensional orthogonal vectors that can best be used to
represent the data, where k <= n.
 The original data are thus projected onto a much smaller
space, resulting in dimensionality reduction.

October 28, 2021 Data Mining: Concepts and Techniques 38

 Numerosity Reduction
 Why numerosity reduction?
 Reduce data volume by choosing alternatives , smaller
forms of representation
 Parametric methods
 Assume the data fits some model,
 estimate model parameters,
 store only the parameters, and discard actual data
 Example :Log-linear models
 Non-parametric methods
 Do not assume models
 Major families: histograms, clustering, sampling

October 28, 2021 Data Mining: Concepts and Techniques 39

 Regress Analysis and Log-
Linear Models
 Linear regression: Y =  +  X
 Two parameters ,  and  specify the line and are to be

estimated by using the data at hand.

 where y=response variable and x=predictor variable

 Multiple regression: Y = b0 + b1 X1 + b2 X2.

 where y=response variable and x1,x2=predictor variable,

b0,b1,b2 as constants
 Log-linear models:
 Higher dimensional data space to be constructed from lower

dimensional space
 Lower dimensional points together occupy less space than

original data points

 The multidimensional probabilities is approximated by a

product of lower-order tables.

 Probability: p(a, b, c, d) = ab acad bcd
 Histograms
 One of the best ways to summarize data is to provide a histogram of the
data
 Store data into buckets
 If bucket store only single value-> singleton ,otherwise range histogram

1,1,5,5,5,5,5,8,8,10,10,10,10,12,14,14,14,15,15,15
,
15,15,15,18,18,18,18,18,18,18,18,20,20,20,20,20,
20,20,21,21,21,21,25,25,25,25,25,28,28,30,30,30

count
1-10 11-20 21-30

October 28, 2021 Data Mining: Concepts and Techniques 41

 Histograms

For this sample database

shown, we can create a
histogram of eye color by
counting the number of
occurrences of different
colors of eyes in our
database

October 28, 2021 Data Mining: Concepts and Techniques 42

 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”
 The quality of cluster may be represented by cluster diameter
 Centroid distance is an alternative measure of cluster quality –
distance from any cluster object to its cluster centroid

October 28, 2021 Data Mining: Concepts and Techniques 43

 Sampling
 Sampling allow a large dataset to be represented by a
much smaller random sample(subset) of the data
 Type of sampling :
 Simple random sample without replacement (SRSWOR)
 Simple random sample with replacement (SRSWR)
 Cluster Sample
 Tuples in database D are grouped into M clusters, than SRS
Of s cluster can be obtained where s<M
 Stratified sampling
 Database D is divided into disjoint parts, Strata
 A stratified sample of D is generated by obtaining an SRS on
each stratum

October 28, 2021 Data Mining: Concepts and Techniques 44

 Sampling

W O R
SRS le random
i m p h ou t
( s e wi t
l
samp ment)
pl a c e
re

SRSW
R

Raw Data
October 28, 2021 Data Mining: Concepts and Techniques 45
 Sampling

Raw Data Cluster/Stratified Sample

October 28, 2021 Data Mining: Concepts and Techniques 46

 Hierarchical Reduction
 Use multi-resolution structure with different degrees of
reduction
 Hierarchical clustering is often performed but tends to
define partitions of data sets rather than “clusters”
 Parametric methods are usually not amenable to
hierarchical representation
 Hierarchical aggregation
 An index tree hierarchically divides a data set into

partitions by value range of some attributes

 Each partition can be considered as a bucket

 Thus an index tree with aggregates stored at each

node is a hierarchical histogram

October 28, 2021 Data Mining: Concepts and Techniques 47
 Chapter 3: Data Preprocessing

 Why preprocess the data?

 Data cleaning
 Data integration and transformation
 Data reduction
 Discretization and concept hierarchy generation
 Summary

October 28, 2021 Data Mining: Concepts and Techniques 48

 Discretization
 Three types of attributes:
 Nominal — values from an unordered set

 Ordinal — values from an ordered set

 Continuous — real numbers

 Discretization:
 divide the range of a continuous attribute into

intervals
 Some classification algorithms only accept categorical

attributes.
 Reduce data size by discretization

 Prepare for further analysis

October 28, 2021 Data Mining: Concepts and Techniques 49

 Discretization and Concept hierachy

 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.
 Concept hierarchies
 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).

October 28, 2021 Data Mining: Concepts and Techniques 50

 Discretization and concept hierarchy
generation for numeric data

 Binning (see sections before)

 Histogram analysis (see sections before)

 Clustering analysis (see sections before)

 Entropy-based discretization

 Segmentation by natural partitioning

October 28, 2021 Data Mining: Concepts and Techniques 51

 Binning
 Attribute values (for one attribute e.g., age):
 0, 4, 12, 16, 16, 18, 24, 26, 28
 Equi-width binning – for bin width of e.g., 10:
 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
 Equi-frequency binning – for bin density of e.g., 3:
 Bin 1: 0, 4, 12 [-, 14) bin
 Bin 2: 16, 16, 18 [14, 21) bin
 Bin 3: 24, 26, 28 [21,+] bin

CS583, Bing Liu, UIC 52

 Entropy-Based Discretization
 Given a set of samples S, if S is partitioned into two
intervals S1 and S2 using boundary T, the entropy after
partitioning is |S | |S |
E (S , T )  1 Ent ( )  2 Ent ( )
|S| S1 | S | S2
 The boundary that minimizes the entropy function over all
possible boundaries is selected as a binary discretization.
 The process is recursively applied to partitions obtained
until some stopping criterion is met, e.g.,
Ent ( S )  E (T , S )  
 Experiments show that it may reduce data size and
improve classification accuracy

October 28, 2021 Data Mining: Concepts and Techniques 53

 Segmentation by natural partitioning

3-4-5 rule can be used to segment numeric data into

relatively uniform, “natural” intervals.
* If an interval covers 3, 6, 7 or 9 distinct values at the
most significant digit, partition the range into 3 equi-
width intervals
* If it covers 2, 4, or 8 distinct values at the most
significant digit, partition the range into 4 intervals
* If it covers 1, 5, or 10 distinct values at the most
significant digit, partition the range into 5 intervals
October 28, 2021 Data Mining: Concepts and Techniques 54
 Example of 3-4-5 rule
count

Step 1: -$351 -$159 profit $1,838 $4,700

Min Low (i.e, 5%-tile) High(i.e, 95%-0 tile) Max
Step 2: msd=1,000 Low=-$1,000 High=$2,000

(-$1,000 - $2,000)
Step 3:

(-$1,000 - 0) (0 -$ 1,000) ($1,000 - $2,000)

(-$4000 -$5,000)
Step 4:

($2,000 - $5, 000)

(-$400 - 0) (0 - $1,000) ($1,000 - $2, 000)
(0 -
($1,000 -
(-$400 - $200)
$1,200) ($2,000 -
-$300) $3,000)
($200 -
($1,200 -
$400)
(-$300 - $1,400)
($3,000 -
-$200)
($400 - ($1,400 - $4,000)
(-$200 - $600) $1,600) ($4,000 -
-$100) ($600 - ($1,600 - $5,000)
$800) ($800 - ($1,800 -
$1,800)
(-$100 - $1,000) $2,000)
0)
October 28, 2021 Data Mining: Concepts and Techniques 55
 Concept hierarchy generation for
categorical data

 Specification of a partial ordering of attributes explicitly

at the schema level by users or experts
 Specification of a portion of a hierarchy by explicit data
grouping
 Specification of a set of attributes, but not of their
partial ordering
 Specification of only a partial set of attributes

October 28, 2021 Data Mining: Concepts and Techniques 56

 Specification of a set of attributes
Concept hierarchy can be automatically generated based
on the number of distinct values per attribute in the
given attribute set. The attribute with the most
distinct values is placed at the lowest level of the
hierarchy.

country 15 distinct values

province_or_ state 65 distinct

values
city 3567 distinct values

street 674,339 distinct values

October 28, 2021 Data Mining: Concepts and Techniques 57
 Chapter 3: Data Preprocessing

 Why preprocess the data?

 Data cleaning
 Data integration and transformation
 Data reduction
 Discretization and concept hierarchy generation
 Summary

October 28, 2021 Data Mining: Concepts and Techniques 58

 Summary

 Data preparation is a big issue for both warehousing

and mining
 Data preparation includes
 Data cleaning and data integration
 Data reduction and feature selection
 Discretization
 A lot a methods have been developed but still an active
area of research

October 28, 2021 Data Mining: Concepts and Techniques 59