HIT391

MACHINE LEARNING:
ADVANCEMENTS
AND APPLICATIONS
 Lecturer: Dr. Yan Zhang

 Email: [email protected]

 Office: Purple 12.3.4

 Week 3:
Data Analysis and Data Pre-processing

 Learning Outcomes

- Data analysis
- Data preprocessing
 An example of Supervised Learning

Data
��
ML algrithm
Training
Classification
Data
�� ��
Step 1: Training Algorithms

Learning from data

Classifier �
(Model)
Learned
model

Step2: Testing
(<40, high, yes, fair) Classifier Buy
��
(Model) computer?
Prediction
 Outline
 An Overview
 Data Cleaning
 Data Integration
 Data Reduction
 Data Transformation and Data
Discretization
 Summary
An Overview
 Data Quality: Why Preprocess the Data?

 Measures for data quality: a multidimensional view

– Accuracy: correct or wrong, accurate or not, e.g.,
Synthetic data
– Completeness: not recorded, unavailable, …e.g., missing
data.
– Consistency: some modified but some not, dangling, …
– Timeliness: timely update? e.g., collected 10 years ago
– Believability: how trustable the data are correct?
– Interpretability: how easily the data can be understood?
 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 reduction
– Dimensionality reduction
– Numerosity reduction
– Data compression
 Data transformation and data discretization
– Normalization. e.g., all data need to be normalized a range [0,1]
– Concept hierarchy generation
Data Cleaning
 Data Cleaning
 Data in the Real World Is Dirty: Lots of potentially incorrect data,
e.g., instrument faulty, human or computer error, transmission error
– incomplete: lacking attribute values, lacking certain attributes of
interest, or containing only aggregate data
 e.g., Occupation=“ ” (missing data)
– noisy: containing noise, errors, or outliers
 e.g., Salary=“−10” (an error)
– inconsistent: containing discrepancies in codes or names, e.g.,
 Age=“42”, Birthday=“03/07/2010”
 Was rating “1, 2, 3”, now rating “A, B, C”
– Intentional (e.g., disguised missing data)
 Jan. 1 as everyone’s birthday?
 Incomplete (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, e.g., recommeder
system
 How to Handle Missing Data?
 Ignore the tuple: usually done when class label is
missing (when doing classification)—not effective
when the % of missing values per attribute varies
considerably
 Fill in the missing value manually: tedious + infeasible?
 Fill in it automatically with
– a global constant: e.g., “unknown”, a new class?!
– the attribute mean
– the attribute mean for all samples belonging to the
same class: smarter
 Noisy Data
 Noise: random error or variance in a measured variable

 Incorrect attribute values may be due to

– faulty data collection instruments

– data entry problems

– data transmission problems

– technology limitation

– inconsistency in naming convention

 Other data problems which require data cleaning

– duplicate records

– incomplete data

– inconsistent data
 How to Handle Noisy Data?
 Binning

– first sort data and partition into (equal-frequency) bins

– then one can smooth by bin means, smooth by bin median, smooth by bin
boundaries, etc.

 Regression

– smooth by fitting the data into regression functions

 Clustering

– detect and remove outliers

 Combined computer and human inspection

– detect suspicious values and check by human (e.g., deal with possible
outliers)
Data Integration
 Data integration
 Data integration:
– Combines data from multiple sources into a coherent store

 Schema integration: e.g., A.cust-id , B.cust-#

– Integrate metadata from different sources

 Entity identification problem:

– Identify real world entities from multiple data sources, e.g., Bill Clinton =
William Clinton

 Detecting and resolving data value conflicts

– For the same real world entity, attribute values from different sources are
different
– Possible reasons: different representations, different scales, e.g., metric
vs. British units
 Handling Redundancy
 Redundant data occur often when integration of
multiple databases
– Object identification: The same attribute or object
may have different names in different databases
– Derivable data: One attribute may be a “derived”
attribute in another table, e.g., annual revenue
 Redundant attributes may be able to be detected by
correlation analysis and covariance analysis
 Careful integration of the data from multiple sources
may help reduce/avoid redundancies and
inconsistencies and improve mining speed and quality
 Correlation Analysis (Numeric Data)
 Correlation coefficient (also called Pearson’s product
moment coefficient)

where n is the number of tuples, and are the

respective means of A and B, σA and σB are the
respective standard deviation of A and B, and
Σ(aibi) is the sum of the AB cross-product.
 If rA,B > 0, A and B are positively correlated (A’s
values increase as B’s). The higher, the stronger
correlation.
 rA,B = 0: independent; rAB < 0: negatively correlated
 Covariance (Numeric Data)
 Covariance is similar to correlation

Correlation coefficient:

where n is the number of tuples, and are the respective mean

or expected values of A and B, σA and σB are the respective
standard deviation of A and B.
 Positive covariance: If CovA,B > 0, then A and B both tend to be
larger than their expected values.
 Negative covariance: If CovA,B < 0 then if A is larger than its
expected value, B is likely to be smaller than its expected value.
 Independence: CovA,B = 0 but the converse is not true。

19
 Co-Variance: An Example

 It can be simplified in computation as

 Suppose two stocks A and B have the following values in one

week: (2, 5), (3, 8), (5, 10), (4, 11), (6, 14).

 Question: If the stocks are affected by the same industry trends,

will their prices rise or fall together?

 E(A) = (2 + 3 + 5 + 4 + 6)/ 5 = 20/5 = 4

 E(B) = (5 + 8 + 10 + 11 + 14) /5 = 48/5 = 9.6

 Cov(A,B) = (2×5+3×8+5×10+4×11+6×14)/5 − 4 × 9.6 = 4

 Thus, A and B rise together since Cov(A, B) > 0.

Data Reduction
 Data Reduction Methods
 Data reduction: Obtain a reduced representation of the data set
that is much smaller in volume but yet produces the same (or
almost the same) analytical results
 Why data reduction? — A database/data warehouse may store
terabytes of data. Complex data analysis may take a very long
time to run on the complete data set.
 Data reduction strategies
– Dimensionality reduction, e.g., remove unimportant attributes
 Principal Components Analysis (PCA)
 Feature subset selection, feature creation
– Data reduction
 Regression and Log-Linear Models
 Histograms, clustering, sampling
 Data cube aggregation
– Data compression
 Dimensionality Reduction
 Curse of dimensionality
– When dimensionality increases, data becomes increasingly sparse
– Density and distance between points, which is critical to clustering, outlier
analysis, becomes less meaningful
– The possible combinations of subspaces will grow exponentially
 Dimensionality reduction
– Avoid the curse of dimensionality
– Help eliminate irrelevant features and reduce noise
– Reduce time and space required in data mining
– Allow easier visualization
 Dimensionality reduction techniques
– Principal Component Analysis
– Supervised and nonlinear techniques (e.g., feature selection)
 Principal Component Analysis (PCA)
 Find a projection that captures the largest amount of variation in
data
 The original data are projected onto a much smaller space,
resulting in dimensionality reduction. We find the eigenvectors of
the covariance matrix, and these eigenvectors define the new
space
x2

x1
 Steps of PCA
 Given N data vectors from n-dimensions, find k ≤ n orthogonal vectors
(principal components) that can be best used to represent data
– Normalize input data: Each attribute falls within the same range
– Compute k orthonormal (unit) vectors, i.e., principal components
– Each input data (vector) is a linear combination of the k principal
component vectors
– The principal components are sorted in order of decreasing
“significance” or strength
– Since the components are sorted, the size of the data can be reduced
by eliminating the weak components, i.e., those with low variance (i.e.,
using the strongest principal components, it is possible to reconstruct
a good approximation of the original data)
 Works for numeric data only
 Attribute Subset Selection
 Another way to reduce dimensionality of data
 Redundant attributes
– Duplicate much or all of the information contained in
one or more other attributes
E.g., purchase price of a product and the amount of
sales tax paid
 Irrelevant attributes
– Contain no information that is useful for the data
mining task at hand
– E.g., students' ID is often irrelevant to the task of
predicting students' GPA
 Data Reduction
 Reduce data volume by choosing alternative, smaller forms of data
representation

 Parametric methods (e.g., regression)

– Assume the data fits some model, estimate model parameters, store only
the parameters, and discard the data (except possible outliers)

– Ex.: Log-linear models—obtain value at a point in m-D space as the

product on appropriate marginal subspaces

 Non-parametric methods

– Do not assume models

– Major families: histograms, clustering, sampling, …

 Regression and Log-Linear Models

 Linear regression
– Data modeled to fit a straight line
– Often uses the least-square method to fit the line

 Multiple regression
– Allows a response variable Y to be modeled as a
linear function of multidimensional feature vector

 Log-linear model
– Approximates discrete multidimensional probability
distributions
 y
Regression Analysis
Y2
 Regression analysis: A collective name
for techniques for the modeling and
analysis of numerical data consisting Y1 y=x+1
of values of a dependent variable (also
called response variable or
x
measurement) and of one or more X1
independent variables (aka.
explanatory variables or predictors)  Used for prediction
 The parameters are estimated so as to (including forecasting of
time-series data),
give a "best fit" of the data
inference, hypothesis
 Most commonly the best fit is testing, and modeling of
evaluated by using the least squares causal relationships
method, but other criteria have also
been used 29
 Clustering
 Partition data set into clusters based
on similarity, and store cluster
representation (e.g., centroid and
diameter) only
 Can be very effective if data is
clustered but not if data is “smeared”
 Can have hierarchical clustering and
be stored in multi-dimensional index
tree structures
 There are many choices of clustering
definitions and clustering algorithms
 Sampling
 Sampling: obtaining a small sample s to represent the
whole data set N
 Allow a mining algorithm to run in complexity that is
potentially sub-linear to the size of the data
 Key principle: Choose a representative subset of the
data
– Simple random sampling may have very poor
performance in the presence of skew
– Develop adaptive sampling methods, e.g., stratified
sampling:
 Note: Sampling may not reduce database I/Os (page
at a time)
 Types of Sampling
 Simple random sampling
– There is an equal probability of selecting any
particular item
 Sampling without replacement
– Once an object is selected, it is removed from the
population
 Sampling with replacement
– A selected object is not removed from the population
 Stratified sampling:
– Partition the data set, and draw samples from each
partition (proportionally, i.e., approximately the same
percentage of the data)
– Used in conjunction with skewed data
 Stratified Sampling

Raw Data Cluster/Stratified Sample

 Data Compression
 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

 Time sequence is not audio

– Typically short and vary slowly with time

 Dimensionality and numerosity reduction may also be considered as

forms of data compression
 Data Compression

lossy
Original Data Compressed
Data
lossless

o ssy
l
Original Data
Approximated
Data Transformation
 Data Transformation
 A function that maps the entire set of values of a given attribute to a new set of replacement
values s.t. each old value can be identified with one of the new values

 Methods

– Smoothing: Remove noise from data

– Attribute/feature construction

 New attributes constructed from the given ones

– Aggregation: Summarization

– Normalization: Scaled to fall within a smaller, specified range

 min-max normalization

 z-score normalization

 normalization by decimal scaling

– Discretization: one-hot encoding, Concept hierarchy climbing

 Normalization

 Min-max normalization: to [new_minA, new_maxA]

– Ex. Let income range $12,000 to $98,000 normalized to [0.0,

1.0]. Then $73,000 is mapped to
 Z-score normalization (μ: mean, σ: standard deviation):

– Ex. Let μ = 54,000, σ = 16,000. Then

 Normalization by decimal scaling
Where j is the smallest integer such that Max(|ν’|) < 1

38
 Discretization
 Three types of attributes
– Nominal—values from an unordered set, e.g., color, profession
– Ordinal—values from an ordered set, e.g., military or academic rank
– Numeric—real numbers, e.g., integer or real numbers

 Discretization: Divide the range of a continuous attribute into intervals

– Interval labels can then be used to replace actual data values
– Reduce data size by discretization
– Supervised vs. unsupervised
– Split (top-down) vs. merge (bottom-up)
– Discretization can be performed recursively on an attribute
– Prepare for further analysis, e.g., classification
 Data Discretization Methods
 Typical methods: All the methods can be applied
recursively
– Binning
 Top-down split, unsupervised
– Clustering analysis (unsupervised, top-down split or
bottom-up merge)
– Decision-tree analysis (supervised, top-down split)
– Correlation analysis (unsupervised, bottom-up
merge)
 Discretization by Classification &
Correlation Analysis
 Concept hierarchy organizes concepts (i.e., attribute values)
hierarchically and is usually associated with each dimension in a
data warehouse
 Concept hierarchies facilitate drilling and rolling in data
warehouses to view data in multiple granularity
 Concept hierarchy formation: Recursively reduce the data by
collecting and replacing low level concepts (such as numeric
values for age) by higher level concepts (such as youth, adult, or
senior)
 Concept hierarchies can be explicitly specified by domain experts
and/or data warehouse designers
 Concept hierarchy can be automatically formed for both numeric
and nominal data. For numeric data, use discretization methods
shown.
 Summary
 Data quality: accuracy, completeness, consistency, timeliness, believability, interpretability

 Data cleaning: e.g. missing/noisy values, outliers

 Data integration from multiple sources:

– Entity identification problem

– Remove redundancies

– Detect inconsistencies

 Data reduction

– Dimensionality reduction

– Numerosity reduction

– Data compression

 Data transformation and data discretization

– Normalization

– Concept hierarchy generation

 References
 Han, Jiawei, Jian Pei, and Hanghang Tong. Data
mining: concepts and techniques. Morgan
kaufmann, 2022.

 Chapter 3 - Concepts and Techniques, Jiawei

Han, Micheline Kamber, and Jian Pei, University
of Illinois at Urbana-Champaign & Simon Fraser
University ©2011 Han, Kamber & Pei.