Data Mining:

Concepts and Techniques

(3rd ed.)

— Chapter 8 —

1
 Chapter 8. Classification: Basic
Concepts
■ Classification: Basic Concepts
■ Decision Tree Induction
■ Bayes Classification Methods
■ Rule-Based Classification
■ Model Evaluation and Selection
■ Techniques to Improve Classification Accuracy:
Ensemble Methods
■ Summary
2
 What is Classification

■ A bank loans officer needs analysis of her data to

learn which loan applicants are “safe” and which are
“risky” for the bank.
■ A marketing manager at AllElectronics needs data
analysis to help guess whether a customer with a
given profile will buy a new computer. (Yes/No)
■ A medical researcher wants to analyze cancer data
to predict which one of three specific treatments a
patient should receive. (A/B/C)
■ In each of these examples, the data analysis task is
classification, where a model or classifier is
constructed to predict class (categorical) labels, 3
 What is Prediction

■ Suppose that the marketing manager wants to

predict how much a given customer will spend
during a sale at AllElectronics.
■ This data analysis task is an example of numeric
prediction, where the model constructed predicts a
continuous-valued function, or ordered value, as
opposed to a class label.
■ This model is a predictor. Regression analysis is a
statistical methodology that is most often used for
numeric prediction

4
 Classification—A Two-Step
Process
■ Model construction: describing a set of predetermined classes
■ Each tuple/sample is assumed to belong to a predefined class, as
determined by the class label attribute
■ The set of tuples used for model construction is training set
■ The model is represented as classification rules, decision trees, or
mathematical formula.
■ Model usage: for classifying future or unknown objects
■ Estimate accuracy of the model
■ The known label of test sample is compared with the classified
result from the model
■ Accuracy rate is the percentage of test set samples that are
correctly classified by the model
■ Test set is independent of training set (otherwise overfitting)
■ If the accuracy is acceptable, use the model to classify data tuples
whose class labels are not known

5
Learning and model construction

6
 Terminology

■ Training dataset
■ Attribute vector
■ Class label attribute
■ Training sample/example/instance/object

7
 Test and Classification

■ Classification: Test data are used to estimate the accuracy of the

classification rules. If the accuracy is considered acceptable, the rules can be
applied to the classification of new data tuples. 8
 Terminology

■ Test dataset
■ Test samples
■ Accuracy of the model

9
Process (1): Model Construction

Classificati
on
Training Algorithms
Data

Classifi
er
(Model)

IF rank =
‘professor’
OR years > 6
THEN tenured =
‘yes’ 10
Process (2): Using the Model in
Prediction

Classifi
er

Testing Unseen
Data Data

(Jeff, Professor,
4)
Tenure
d?

11
 Evaluation Criteria
■ Accuracy on test set
■ the rate of correct
classification on the testing
set. E.g., if 90 are classified
correctly out of the 100
testing cases, accuracy is 90%. Predicted class
■ Error Rate on test set
■ The percentage of wrong
predictions on test set Yes No
■ Confusion Matrix
■ For binary class values, “yes”
and “no”, a matrix showing Actual Yes True False
true positive, true negative,
false positive and false class positive negativ
negative rates e
■ Speed and scalability
■ the time to build the classifier
and to classify new cases, and No False True
the scalability with respect to
the data size. positive negativ
■ Robustness: handling noise e
and missing values
 Evaluation Techniques

■ Holdout: the training set/testing set.

■ Good for a large set of data.

■ k-fold Cross-validation:
■ divide the data set into k sub-samples.

■ In each run, use one distinct sub-sample as testing

set and the remaining k-1 sub-samples as training

set.
■ Evaluate the method using the average of the k
runs.
■ This method reduces the randomness of training set/
testing set.
 Cross Validation: Holdout Method
— Break up data into groups of the same size
—

— Hold aside one group for testing and use the rest to build model

— Repeat

iteration
Test

14
14
Supervised vs. Unsupervised Learning
■ Supervised learning (classification)
■ Supervision: The training data (observations,
measurements, etc.) are accompanied by labels indicating
the class of the observations
■ New data is classified based on the training set
■ Unsupervised learning (clustering)
■ The class labels of training data is unknown
■ Given a set of measurements, observations, etc. with the
aim of establishing the existence of classes or clusters in
the data
15
Prediction Problems: Classification vs.
Numeric Prediction
■ Classification
■ predicts categorical class labels (discrete or nominal)

■ classifies data (constructs a model) based on the training

set and the values (class labels) in a classifying attribute
and uses it in classifying new data
■ Numeric Prediction
■ models continuous-valued functions, i.e., predicts
unknown or missing values
■ Typical applications
■ Credit/loan approval:

■ Medical diagnosis: if a tumor is cancerous or benign

■ Fraud detection: if a transaction is fraudulent

■ Web page categorization: which category it is

16
 Chapter 8. Classification: Basic
Concepts
■ Classification: Basic Concepts
■ Decision Tree Induction
■ Bayes Classification Methods
■ Rule-Based Classification
■ Model Evaluation and Selection
■ Techniques to Improve Classification Accuracy:
Ensemble Methods
■ Summary
17
 Terminology

■ Decision tree induction is the learning of decision

trees from class-labeled training tuples.
■ A decision tree is a flowchart-like tree structure,
■ where each internal node (nonleaf node) denotes a
test on an attribute,
■ Each branch represents an outcome of the test,
■ and each leaf node (or terminal node) holds a class
label.
■ The topmost node in a tree is the root node.

18
 Decision Tree Induction: An Example
❑ Training data set: Buys_computer
❑ The data set follows an example of
Quinlan’s ID3 (Playing Tennis)
❑ Resulting tree:
age
?
<=3 overca
31..40 >4
0 st 0
studen ye credit
t? s rating?
no ye excelle fai
s nt r
n ye n ye
o s 19
 Why decision tree
■ The construction of decision tree classifiers does not require
any domain knowledge or parameter setting, and therefore is
appropriate for exploratory knowledge discovery.
■ Decision trees can handle multidimensional data. Their
representation of acquired knowledge in tree form is intuitive
and generally easy to assimilate by humans.
■ The learning and classification steps of decision tree induction
are simple and fast. In general, decision tree classifiers have
good accuracy. However, successful use may depend on the
data at hand. Decision tree induction algorithms have been
used for classification in many application areas such as
medicine, manufacturing and production, financial analysis,
astronomy, and molecular biology. Decision trees are the basis
of several commercial rule induction systems.
20
 Concepts in leaning decision tree
■ Attribute selection measures are used to select the attribute
that best partitions the tuples into distinct classes.
■ When decision trees are built, many of the branches may
reflect noise or outliers in the training data. Tree pruning
attempts to identify and remove such branches, with the goal
of improving classification accuracy on unseen data.
■ Scalability is a big issues for the induction of decision trees
from large databases

21
 Tree algorithms

■ ID3 (Iterative Dichotomiser): J. Ross Quinlan, a researcher in

machine learning, developed a decision tree algorithm
■ C4.5(a successor of ID3)
■ CART(Classification and Regression Trees )

22
Algorithm for Decision Tree Induction
■ Basic algorithm (a greedy algorithm)
■ Tree is constructed in a top-down recursive divide-and-

conquer manner
■ At start, all the training examples are at the root

■ Attributes are categorical (if continuous-valued, they are

discretized in advance)
■ Examples are partitioned recursively based on selected

attributes
■ Test attributes are selected on the basis of a heuristic or

statistical measure (e.g., information gain)

■ Conditions for stopping partitioning
■ All samples for a given node belong to the same class

■ There are no remaining attributes for further partitioning –

majority voting is employed for classifying the leaf

■ There are no samples left
23
 Choosing the Splitting Attribute

■ At each node, available attributes are evaluated on

the basis of separating the classes of the training
examples. A Goodness function is used for this
purpose.
■ Typical goodness functions:
■ information gain (ID3/C4.5)

■ information gain ratio

■ gini index

24
 Attribute Selection Measure:
Information Gain (ID3/C4.5)
■ Select the attribute with the highest information gain
■ Let pi be the probability that an arbitrary tuple in D belongs to
class Ci, estimated by |C i, D|/|D|
■ Expected information (entropy) needed to classify a tuple in D:

■ Information needed (after using A to split D into v partitions) to

classify D:

■ Information gained by branching on attribute A

25
 Attribute Selection: Information Gain
g Class P: buys_computer = “yes”
g Class N: buys_computer = “no”

26
 Attribute Selection: Information Gain
g Class P: buys_computer = “yes”
g Class N: buys_computer = “no”

means “age <=30” has 5 out of

14 samples, with 2 yes’es and 3
no’s.

27
 Attribute Selection: Information Gain
g Class P: buys_computer = “yes”
g Class N: buys_computer = “no”

28
29
30
■ Conditions for stopping
partitioning
■ All samples for a given node

belong to the same class

■ There are no remaining

attributes for further

partitioning – majority voting
is employed for classifying
the leaf
■ There are no samples left

31
32
 Computing Information-Gain for
Continuous-Valued Attributes
■ Let attribute A be a continuous-valued attribute
■ Must determine the best split point for A
■ Sort the value A in increasing order
■ Typically, the midpoint between each pair of adjacent values
is considered as a possible split point
■ (a i+ai+1)/2 is the midpoint between the values of a i and a i+1
■ The point with the minimum expected information
requirement for A is selected as the split-point for A
■ Split:
■ D1 is the set of tuples in D satisfying A ≤ split-point, and D2 is
the set of tuples in D satisfying A > split-point
33
34
 Highly-branching attributes

■ Problematic: attributes with a large number of values

(extreme case: ID code)
■ Subsets are more likely to be pure if there is a large
number of values
⇒ Information gain is biased towards choosing

attributes with a large number of values

⇒ This may result in overfitting (selection of an

attribute that is non-optimal for prediction)

■ Another problem: fragmentation

35
 The gain ratio
■ Gain ratio: a modification of the information gain that
reduces its bias on high-branch attributes
■ Gain ratio takes number and size of branches into
account when choosing an attribute
■ It corrects the information gain by taking the
intrinsic information of a split into account
■ Also called split ratio

■ Intrinsic information: entropy of distribution of

instances into branches
■ (i.e. how much info do we need to tell which
branch an instance belongs to)

36
 Gain Ratio for Attribute Selection
(C4.5)
■ Information gain measure is biased towards attributes with a
large number of values
■ C4.5 (a successor of ID3) uses gain ratio to overcome the
problem (normalization to information gain)

■ GainRatio(A) = Gain(A)/SplitInfo(A)
■ Ex.

■ gain_ratio(income) = 0.029/1.557 = 0.019

■ The attribute with the maximum gain ratio is selected as the
splitting attribute
37
38
 Gini Index (CART, IBM
IntelligentMiner)
■ If a data set D contains examples from n classes, gini index,
gini(D) is defined as

where pj is the relative frequency of class j in D

■ If a data set D is split on A into two subsets D1 and D2, the gini
index gini(D) is defined as

■ Reduction in Impurity:

■ The attribute provides the smallest ginisplit(D) (or the largest

reduction in impurity) is chosen to split the node (need to
enumerate all the possible splitting points for each attribute)

39
40
41
42
43
44
45
 Computation of Gini Index
■ Ex. D has 9 tuples in buys_computer = “yes” and 5 in “no”

■ Suppose the attribute income partitions D into 10 in D1: {low,

medium} and 4 in D 2

Gini{low,high} is 0.458; Gini{medium,high} is 0.450. Thus, split on the

{low,medium} (and {high}) since it has the lowest Gini index

46
Yes and No are incorrect here, add them according to data table

47
48
49
50
 Comparing Attribute Selection Measures

■ The three measures, in general, return good results but

■ Information gain:

■ biased towards multivalued attributes

■ Gain ratio:
■ tends to prefer unbalanced splits in which one partition
is much smaller than the others
■ Gini index:
■ biased to multivalued attributes
■ has difficulty when # of classes is large
■ tends to favor tests that result in equal-sized partitions
and purity in both partitions
51
 Other Attribute Selection Measures
■ CHAID: a popular decision tree algorithm, measure based on χ 2 test for
independence
■ C-SEP: performs better than info. gain and gini index in certain cases
■ G-statistic: has a close approximation to χ 2 distribution
■ MDL (Minimal Description Length) principle (i.e., the simplest solution is
preferred):
■ The best tree as the one that requires the fewest # of bits to both (1)
encode the tree, and (2) encode the exceptions to the tree
■ Multivariate splits (partition based on multiple variable combinations)
■ CART: finds multivariate splits based on a linear comb. of attrs.
■ Which attribute selection measure is the best?
■ Most give good results, none is significantly superior than others
52
ZeroR is the simplest classification method
which relies on the target and ignores all
predictors. ZeroR classifier simply predicts
the majority category (class). Although
there is no predictability power in ZeroR,
it is useful for determining a baseline
performance as a benchmark for other
classification methods

53
 Overfitting and Tree Pruning
■ Overfitting: An induced tree may overfit the training data
■ Too many branches, some may reflect anomalies due to

noise or outliers
■ Poor accuracy for unseen samples

■ Two approaches to avoid overfitting

■ Prepruning: Halt tree construction early ̵ do not split a node

if this would result in the goodness measure falling below a

threshold
■ Difficult to choose an appropriate threshold
■ Postpruning: Remove branches from a “fully grown”

tree—get a sequence of progressively pruned trees

■ Use a set of data different from the training data to
decide which is the “best pruned tree”
54
 Enhancements to Basic Decision Tree
Induction
■ Allow for continuous-valued attributes
■ Dynamically define new discrete-valued attributes that

partition the continuous attribute value into a discrete set of

intervals
■ Handle missing attribute values
■ Assign the most common value of the attribute
■ Assign probability to each of the possible values

■ Attribute construction
■ Create new attributes based on existing ones that are

sparsely represented
■ This reduces fragmentation, repetition, and replication

55
 Classification in Large Databases

■ Classification—a classical problem extensively studied by statisticians and

machine learning researchers
■ Scalability: Classifying data sets with millions of examples and hundreds of
attributes with reasonable speed
■ Why is decision tree induction popular?
■ relatively faster learning speed (than other classification methods)
■ convertible to simple and easy to understand classification rules
■ can use SQL queries for accessing databases
■ comparable classification accuracy with other methods
■ RainForest (VLDB’98 — Gehrke, Ramakrishnan & Ganti)
■ In data mining applications, very large training sets of millions of tuples are
common. Most often, the training data will not fit in memory
■ RainForest, for example, adapts to the amount of main memory available and
applies to any decision tree induction algorithm
■ Builds an AVC-list (attribute, value, class label)

56
 Scalability Framework for
RainForest

■ Separates the scalability aspects from the criteria that

determine the quality of the tree
■ Builds an AVC-list: AVC (Attribute, Value, Class_label)
■ AVC-set (of an attribute X )
■ Projection of training dataset onto the attribute X and
class label where counts of individual class label are
aggregated
■ AVC-group (of a node n )
■ Set of AVC-sets of all predictor attributes at the node n

57
Rainforest: Training Set and Its AVC
Sets

Training Examples AVC-set on AVC-set on

AgeAge Buy_Computer income
income Buy_Computer

yes no
yes no
high 2 2
<=30 2 3
31..40 4 0 medium 4 2

>40 3 2 low 3 1

AVC-set on
AVC-set on
credit_ratin
Student
student Buy_Computer g Buy_Computer
Credit
yes no rating yes no

yes 6 1 fair 6 2

no 3 4 excellent 3 3

58
 BOAT (Bootstrapped Optimistic
Algorithm for Tree Construction)
■ Use a statistical technique called bootstrapping to create
several smaller samples (subsets), each fits in memory
■ Each subset is used to create a tree, resulting in several
trees
■ These trees are examined and used to construct a new tree
T’
■ It turns out that T’ is very close to the tree that would be
generated using the whole data set together
■ Adv: requires only two scans of DB, an incremental alg.
■ BOAT was found to be two to three times faster than
RainForest 59
60
That is All for today!

See you next week!

61
 Chapter 8. Classification: Basic
Concepts
■ Classification: Basic Concepts
■ Decision Tree Induction
■ Bayes Classification Methods
■ Rule-Based Classification
■ Model Evaluation and Selection
■ Techniques to Improve Classification Accuracy:
Ensemble Methods
■ Summary
62
 Bayesian Classification: Why?
■ A statistical classifier: performs probabilistic prediction, i.e., predicts class
membership probabilities
■ Foundation: Based on Bayes’ Theorem.
■ Performance: A simple Bayesian classifier, naïve Bayesian classifier, has
comparable performance with decision tree and selected neural network
classifiers. Bayesian classifiers have also exhibited high accuracy and speed
when applied to large databases.
■ Incremental: Each training example can incrementally increase/decrease the
probability that a hypothesis is correct — prior knowledge can be combined
with observed data
■ Na¨ıve Bayesian classifiers assume that the effect of an attribute value on a
given class is independent of the values of the other attributes. This assumption
is called class conditional independence. It is made to simplify the
computations involved and, in this sense, is considered “na¨ıve.”
■ Standard: Even when Bayesian methods are computationally intractable, they
can provide a standard of optimal decision making against which other methods
can be measured. 63
 Bayesian Theorem: Basics

■ Let X be a data sample (“evidence”)

■ Let H be a hypothesis that X belongs to class C
■ Classification is to determine P(H|X), (posteriori probability),
the probability that the hypothesis holds given the observed
data sample X
■ P(H) (prior probability), the initial probability
■ E.g., X will buy computer, regardless of age, income, …

■ P(X): probability that sample data is observed

■ P(X|H) (likelyhood), the probability of observing the sample X,
given that the hypothesis holds
■ E.g., Given that X will buy computer, the prob. that X is

31..40, medium income

64
 Bayesian Theorem

■ Given training data X, posteriori probability of a hypothesis H,

P(H|X), follows the Bayes theorem

■ Informally, this can be written as

posteriori = likelihood x prior/evidence
■ Predicts X belongs to C 2 iff the probability P(Ci|X) is the highest
among all the P(C k|X) for all the k classes
■ Practical difficulty: require initial knowledge of many
probabilities, significant computational cost

65
 Towards Naïve Bayesian Classifier
■ Let D be a training set of tuples and their associated class
labels, and each tuple is represented by an n-D attribute vector
X = (x 1, x 2, …, x n)
■ Suppose there are m classes C1, C 2, …, C m.
■ Classification is to derive the maximum posteriori, i.e., the
maximal P(Ci|X)
■ This can be derived from Bayes’ theorem

■ Since P(X) is constant for all classes, only

needs to be maximized

66
 Derivation of Naïve Bayes Classifier
■ A simplified assumption: attributes are conditionally
independent (i.e., no dependence relation between
attributes):

■ This greatly reduces the computation cost: Only counts the

class distribution
■ If A k is categorical, P(x k|Ci) is the # of tuples in C i having value
xk for Ak divided by |C i, D | (# of tuples of C i in D)
■ If A k is continous-valued, P(x k|Ci) is usually computed based on
Gaussian distribution with a mean μ and standard deviation σ

and P(xk|Ci) is

67
 Naïve Bayesian Classifier: Training
Dataset

Class:
C1:buys_computer = ‘yes’
C2:buys_computer = ‘no’

Data sample
X = (age <=30,
Income = medium,
Student = yes
Credit_rating = Fair)

68
69
 Avoiding the Zero-Probability
Problem
■ Naïve Bayesian prediction requires each conditional prob. be
non-zero. Otherwise, the predicted prob. will be zero

■ Ex. Suppose a dataset with 1000 tuples, income=low (0),

income= medium (990), and income = high (10)
■ Use Laplacian correction (or Laplacian estimator)
■ Adding 1 to each case

Prob(income = low) = 1/1003

Prob(income = medium) = 991/1003
Prob(income = high) = 11/1003
■ The “corrected” prob. estimates are close to their

“uncorrected” counterparts
70
Naïve Bayesian Classifier: Comments
■ Advantages
■ Easy to implement

■ Good results obtained in most of the cases

■ Disadvantages
■ Assumption: class conditional independence, therefore loss
of accuracy
■ Practically, dependencies exist among variables

■ E.g., hospitals: patients: Profile: age, family history, etc.

Symptoms: fever, cough etc., Disease: lung cancer,
diabetes, etc.
■ Dependencies among these cannot be modeled by Naïve
Bayesian Classifier
■ How to deal with these dependencies? Bayesian Belief Networks
(Chapter 9)
71
 Chapter 8. Classification: Basic
Concepts
■ Classification: Basic Concepts
■ Decision Tree Induction
■ Bayes Classification Methods
■ Rule-Based Classification
■ Model Evaluation and Selection
■ Techniques to Improve Classification Accuracy:
Ensemble Methods
■ Summary
72
 IF-THEN Rules
Rules are a good way of representing information or bits of knowledge. A rule-based
classifier uses a set of IF-THEN rules for classification. An IF-THEN rule is an expression of
the form
IF condition THEN conclusion
Let us consider a rule R1,

R1: IF age = youth AND student = yes THEN buy_computer = yes

Points to remember −

The IF part of the rule is called rule antecedent or precondition.

The THEN part of the rule is called rule consequent.

The antecedent part the condition consist of one or more attribute tests and these tests are
logically ANDed.

The consequent part consists of class prediction.

Note − We can also write rule R1 as follows −

R1: (age = youth) ^ (student = yes))(buys computer = yes)

73
 Rule-Based Classifier

■ Classify records by using a collection of “if…then…”

rules
■ Rule: (Condition) → y
■ where
■ Condition is a conjunctions of attributes
■ y is the class label
■ LHS: rule antecedent or condition
■ RHS: rule consequent
■ Examples of classification rules:
■ (Blood Type=Warm) ∧ (Lay Eggs=Yes) → Birds
■ (Taxable Income < 50K) ∧ (Refund=Yes) → Evade
Rule-based Classifier (Example)

R1: (Give Birth = no) ∧ (Can Fly = yes) → Birds

R2: (Give Birth = no) ∧ (Live in Water = yes) → Fishes
R3: (Give Birth = yes) ∧ (Blood Type = warm) → Mammals
R4: (Give Birth = no) ∧ (Can Fly = no) → Reptiles
R5: (Live in Water = sometimes) → Amphibians
Application of Rule-Based Classifier

■ A rule r covers an instance x if the attributes of the instance

satisfy the condition of the rule
R1: (Give Birth = no) ∧ (Can Fly = yes) → Birds
R2: (Give Birth = no) ∧ (Live in Water = yes) → Fishes
R3: (Give Birth = yes) ∧ (Blood Type = warm) → Mammals
R4: (Give Birth = no) ∧ (Can Fly = no) → Reptiles
R5: (Live in Water = sometimes) → Amphibians

The rule ? covers a hawk => Bird

The rule ? covers the grizzly bear => Mammal
Rule Coverage and Accuracy
Coverage of a rule:
■ Fraction of records that satisfy the
antecedent of a rule
Accuracy of a rule:
■ Fraction of records that satisfy both
the antecedent and consequent of a
rule
❑ Assessment of a rule: coverage
and accuracy
■ ncovers = # of tuples covered by R
■ ncorrect = # of tuples correctly
classified by R
■ coverage(R) = ncovers /|D|
/* D: training data set */
■ accuracy(R) = ncorrect / ncovers
(Status=Single) → No
Coverage = 40%, Accuracy =
50%
How does Rule-based Classifier Work?
R1: (Give Birth = no) ∧ (Can Fly = yes) → Birds
R2: (Give Birth = no) ∧ (Live in Water = yes) → Fishes
R3: (Give Birth = yes) ∧ (Blood Type = warm) → Mammals
R4: (Give Birth = no) ∧ (Can Fly = no) → Reptiles
R5: (Live in Water = sometimes) → Amphibians

A lemur triggers rule R3, so it is classified as a mammal

A turtle triggers both R4 and R5
A dogfish shark triggers none of the rules
Characteristics of Rule-Based
Classifier
Mutually exclusive rules
■ Classifier contains mutually exclusive rules if
the rules are independent of each other
■ Every record is covered by at most one rule

Exhaustive rules
■ Classifier has exhaustive coverage if it
accounts for every possible combination of
attribute values
■ Each record is covered by at least one rule
From Decision Trees To Rules

Rules are mutually exclusive and exhaustive

Rule set contains as much information as
the tree
Rules Can Be Simplified

Initial Rule: (Refund=No) ∧ (Status=Married) → No

Simplified Rule: (Status=Married) → No
Effect of Rule Simplification

Rules are no longer mutually exclusive

■ A record may trigger more than one rule
■ Solution?
Ordered rule set
Unordered rule set – use voting schemes

Rules are no longer exhaustive

■ A record may not trigger any rules
■ Solution?
Use a default class
Ordered Rule Set
If more than one rule are triggered, need conflict resolution
Rules are rank ordered according to their priority
■ An ordered rule set is known as a decision list
When a test record is presented to the classifier
■ It is assigned to the class label of the highest ranked rule
it has triggered
■ If none of the rules fired, it is assigned to the default class

R1: (Give Birth = no) ∧ (Can Fly = yes) → Birds

R2: (Give Birth = no) ∧ (Live in Water = yes) → Fishes
R3: (Give Birth = yes) ∧ (Blood Type = warm) → Mammals
R4: (Give Birth = no) ∧ (Can Fly = no) → Reptiles
R5: (Live in Water = sometimes) → Amphibians
Rule Ordering Schemes
Rule-based ordering
■ Individual rules are ranked based on their quality

Class-based ordering
■ Rules that belong to the same class appear together
Building Classification Rules
Direct Method:
Extract rules directly from data
e.g.: RIPPER, CN2, Holte’s 1R

Indirect Method:
Extract rules from other classification models (e.g.
decision trees, etc).
e.g: C4.5 rules
Direct Method: Sequential Covering
1. Start from an empty rule
2. Grow a rule using the Learn-One-Rule
function
3. Remove training records covered by the
rule
4. Repeat Step (2) and (3) until stopping
criterion is met
Example of Sequential Covering
Example of Sequential Covering…
Aspects of Sequential Covering
Rule Growing

Instance Elimination

Rule Evaluation

Stopping Criterion

Rule Pruning
Rule Growing
Two common
strategies
Rule Growing (Examples)
CN2 Algorithm:
■ Start from an empty conjunct: {}
■ Add conjuncts that minimizes the entropy measure: {A}, {A,B}, …
■ Determine the rule consequent by taking majority class of instances
covered by the rule
RIPPER Algorithm:
■ Start from an empty rule: {} => class
■ Add conjuncts that maximizes FOIL’s information gain measure:
R0: {} => class (initial rule)
R1: {A} => class (rule after adding conjunct)
Gain(R0, R1) = t [ log (p1/(p1+n1)) – log (p0/(p0 + n0)) ]
where t: number of positive instances covered by both R0 and R1
p0: number of positive instances covered by R0
n0: number of negative instances covered by R0
p1: number of positive instances covered by R1
n1: number of negative instances covered by R1
Instance Elimination
Why do we need to
eliminate instances?
■ Otherwise, the next rule
is identical to previous
rule
Why do we remove
positive instances?
■ Ensure that the next rule
is different
Why do we remove
negative instances?
■ Prevent underestimating
accuracy of rule
■ Compare rules R2 and R3
in the diagram
Rule Evaluation
Metrics:
■ Accuracy

n : Total number of
■ Laplace instances covered by rule
nc : Number of instances
covered by rule from n
k : Number of classes
■ M-estimate
p : Prior probability
Stopping Criterion and Rule Pruning
Stopping criterion
■ Compute the gain
■ If gain is not significant, discard the new rule

Rule Pruning
■ Similar to post-pruning of decision trees
■ Reduced Error Pruning:
Remove one of the conjuncts in the rule
Compare error rate on validation set before and after
pruning
If error improves, prune the conjunct
Summary of Direct Method
Grow a single rule

Remove Instances from rule

Prune the rule (if necessary)

Add rule to Current Rule Set

Repeat
 Chapter 8. Classification: Basic
Concepts
■ Classification: Basic Concepts
■ Decision Tree Induction
■ Bayes Classification Methods
■ Rule-Based Classification
■ Model Evaluation and Selection
■ Techniques to Improve Classification Accuracy:
Ensemble Methods
■ Summary
96
 Model Evaluation and Selection
■ Evaluation metrics: How can we measure accuracy? Other
metrics to consider?
■ Use test set of class-labeled tuples instead of training set when
assessing accuracy
■ Methods for estimating a classifier’s accuracy:
■ Holdout method, random subsampling
■ Cross-validation
■ Bootstrap
■ Comparing classifiers:
■ Confidence intervals
■ Cost-benefit analysis and ROC Curves
97
Classifier Evaluation Metrics: Confusion
Matrix
Confusion Matrix:
Actual class\Predicted class C1 ¬ C1
C1 True Positives (TP) False Negatives (FN)
¬ C1 False Positives (FP) True Negatives (TN)

Example of Confusion
Matrix:
Actual buy_computer buy_computer Total
class\Predicted class = yes = no
buy_computer = yes 6954 46 7000
buy_computer = no 412 2588 3000
Total 7366 2634 10000

■ Given m classes, an entry, CMi,j in a confusion matrix indicates

# of tuples in class i that were labeled by the classifier as class j
■ May have extra rows/columns to provide totals
98
 Accuracy, Error Rate, Sensitivity and
Specificity
A\P C ¬C ■ Class Imbalance Problem:
C TP FN P
■ One class may be rare, e.g.
¬C FP TN N
fraud, or HIV-positive
P’ N’ All
■ Significant majority of the

■ Classifier Accuracy, or negative class and minority of

recognition rate: percentage of the positive class
test set tuples that are correctly ■ Sensitivity: True Positive
classified recognition rate
Accuracy = (TP + TN)/All ■ Sensitivity = TP/P
■ Error rate: 1 – accuracy, or ■ Specificity: True Negative

Error rate = (FP + FN)/All recognition rate

■ Specificity = TN/N

99
 Precision and Recall, and F-
measures
■ Precision: exactness – what % of tuples that the classifier
labeled as positive are actually positive

■ Recall: completeness – what % of positive tuples did the

classifier label as positive?
■ Perfect score is 1.0
■ Inverse relationship between precision & recall
■ F measure (F1 or F-score): harmonic mean of precision and
recall,

■ Fß: weighted measure of precision and recall

■ assigns ß times as much weight to recall as to precision

100
Classifier Evaluation Metrics: Example

Actual Class\Predicted class cancer = yes cancer = no Total Recognition(%)

cancer = yes 90 210 300 30.00 (sensitivity
cancer = no 140 9560 9700 98.56 (specificity)
Total 230 9770 10000 96.40 (accuracy)

■ Precision = 90/230 = 39.13% Recall = 90/300 = 30.00%

101
 Holdout & Cross-Validation
Methods
■ Holdout method
■ Given data is randomly partitioned into two independent sets

■ Training set (e.g., 2/3) for model construction

■ Test set (e.g., 1/3) for accuracy estimation
■ Random sampling: a variation of holdout

■ Repeat holdout k times, accuracy = avg. of the accuracies

obtained
■ Cross-validation (k-fold, where k = 10 is most popular)
■ Randomly partition the data into k mutually exclusive subsets,
each approximately equal size
■ At i-th iteration, use D i as test set and others as training set

■ Leave-one-out: k folds where k = # of tuples, for small sized

data
■ *Stratified cross-validation*: folds are stratified so that class
dist. in each fold is approx. the same as that in the initial data
102
 Evaluating Classifier Accuracy:
Bootstrap
■ Bootstrap
■ Works well with small data sets
■ Samples the given training tuples uniformly with replacement
■ i.e., each time a tuple is selected, it is equally likely to be selected
again and re-added to the training set
■ Several bootstrap methods, and a common one is .632 boostrap
■ A data set with d tuples is sampled d times, with replacement, resulting in
a training set of d samples. The data tuples that did not make it into the
training set end up forming the test set. About 63.2% of the original data
end up in the bootstrap, and the remaining 36.8% form the test set (since
(1 – 1/d) d ≈ e-1 = 0.368)
■ Repeat the sampling procedure k times, overall accuracy of the model:

103
 Estimating Confidence Intervals:
Classifier Models M1 vs. M2
■ Suppose we have 2 classifiers, M1 and M 2, which one is better?

■ Use 10-fold cross-validation to obtain and

■ These mean error rates are just estimates of error on the true
population of future data cases
■ What if the difference between the 2 error rates is just
attributed to chance?

■ Use a test of statistical significance

■ Obtain confidence limits for our error estimates

104
 Estimating Confidence Intervals:
Null Hypothesis
■ Perform 10-fold cross-validation
■ Assume samples follow a t distribution with k–1 degrees of
freedom (here, k=10)
■ Use t-test (or Student’s t-test)
■ Null Hypothesis: M 1 & M2 are the same
■ If we can reject null hypothesis, then
■ we conclude that the difference between M 1 & M2 is
statistically significant
■ Chose model with lower error rate

105
 Estimating Confidence Intervals: t-test

■ If only 1 test set available: pairwise comparison

■ For i th round of 10-fold cross-validation, the same cross
partitioning is used to obtain err(M1)i and err(M2)i
■ Average over 10 rounds to get an
d
■ t-test computes t-statistic with k-1 degrees of
freedom: wher
e

■ If two test sets available: use non-paired t-test

wher
e
where k1 & k2 are # of cross-validation samples used for M1 & M2 , resp.
106
 Estimating Confidence Intervals:
Table for t-distribution

■ Symmetric
■ Significance level,
e.g., sig = 0.05 or
5% means M1 & M2
are significantly
different for 95% of
population
■ Confidence limit, z
= sig/2

107
 Estimating Confidence Intervals:
Statistical Significance
■ Are M 1 & M2 significantly different?
■ Compute t. Select significance level (e.g. sig = 5%)

■ Consult table for t-distribution: Find t value corresponding

to k-1 degrees of freedom (here, 9)

■ t-distribution is symmetric: typically upper % points of

distribution shown → look up value for confidence limit

z=sig/2 (here, 0.025)
■ If t > z or t < -z, then t value lies in rejection region:

■ Reject null hypothesis that mean error rates of M 1 & M2

are same
■ Conclude: statistically significant difference between M1
& M2
■ Otherwise, conclude that any difference is chance
108
Model Selection: ROC Curves
■ ROC (Receiver Operating
Characteristics) curves: for visual
comparison of classification models
■ Originated from signal detection theory
■ Shows the trade-off between the true
positive rate and the false positive rate
■ The area under the ROC curve is a ■ Vertical axis
measure of the accuracy of the model represents the true
positive rate
■ Rank the test tuples in decreasing ■ Horizontal axis rep.
order: the one that is most likely to the false positive rate
belong to the positive class appears at ■ The plot also shows
the top of the list a diagonal line
■ The closer to the diagonal line (i.e., the ■ A model with perfect
closer the area is to 0.5), the less accuracy will have an
accurate is the model area of 1.0
109
 Issues Affecting Model Selection

■ Accuracy
■ classifier accuracy: predicting class label

■ Speed
■ time to construct the model (training time)
■ time to use the model (classification/prediction time)
■ Robustness: handling noise and missing values
■ Scalability: efficiency in disk-resident databases
■ Interpretability
■ understanding and insight provided by the model
■ Other measures, e.g., goodness of rules, such as decision tree
size or compactness of classification rules
110
 Chapter 8. Classification: Basic Concepts

■ Classification: Basic Concepts

■ Decision Tree Induction
■ Bayes Classification Methods
■ Rule-Based Classification
■ Model Evaluation and Selection
■ Techniques to Improve Classification Accuracy:
Ensemble Methods
■ Summary
111
 Ensemble Methods: Increasing the
Accuracy

■ Ensemble methods
■ Use a combination of models to increase accuracy

■ Combine a series of k learned models, M 1 , M 2 , …, M k , with

the aim of creating an improved model M*

■ Popular ensemble methods
■ Bagging: averaging the prediction over a collection of

classifiers
■ Boosting: weighted vote with a collection of classifiers

■ Ensemble: combining a set of heterogeneous classifiers

112
 Bagging: Boostrap Aggregation
■ Analogy: Diagnosis based on multiple doctors’ majority vote
■ Training
■ Given a set D of d tuples, at each iteration i, a training set Di of d tuples
is sampled with replacement from D (i.e., bootstrap)
■ A classifier model Mi is learned for each training set D i
■ Classification: classify an unknown sample X
■ Each classifier M i returns its class prediction
■ The bagged classifier M* counts the votes and assigns the class with the
most votes to X
■ Prediction: can be applied to the prediction of continuous values by taking
the average value of each prediction for a given test tuple
■ Accuracy
■ Often significantly better than a single classifier derived from D
■ For noise data: not considerably worse, more robust
■ Proved improved accuracy in prediction
113
 Boosting
■ Analogy: Consult several doctors, based on a combination of
weighted diagnoses—weight assigned based on the previous
diagnosis accuracy
■ How boosting works?
■ Weights are assigned to each training tuple

■ A series of k classifiers is iteratively learned

■ After a classifier M i is learned, the weights are updated to

allow the subsequent classifier, M i+1, to pay more attention
to the training tuples that were misclassified by M i
■ The final M* combines the votes of each individual classifier,
where the weight of each classifier's vote is a function of its
accuracy
■ Boosting algorithm can be extended for numeric prediction
■ Comparing with bagging: Boosting tends to have greater
accuracy, but it also risks overfitting the model to misclassified
114
Adaboost (Freund and Schapire, 1997)
■ Given a set of d class-labeled tuples, (X1 , y 1 ), …, (Xd, y d)
■ Initially, all the weights of tuples are set the same (1/d)
■ Generate k classifiers in k rounds. At round i,
■ Tuples from D are sampled (with replacement) to form a training set
Di of the same size
■ Each tuple’s chance of being selected is based on its weight
■ A classification model M i is derived from D i
■ Its error rate is calculated using Di as a test set
■ If a tuple is misclassified, its weight is increased, o.w. it is decreased
■ Error rate: err(Xj ) is the misclassification error of tuple Xj . Classifier M i
error rate is the sum of the weights of the misclassified tuples:

■ The weight of classifier M i’s vote is

115
 Random Forest (Breiman 2001)
■ Random Forest:
■ Each classifier in the ensemble is a decision tree classifier and is
generated using a random selection of attributes at each node to
determine the split
■ During classification, each tree votes and the most popular class is
returned
■ Two Methods to construct Random Forest:
■ Forest-RI (random input selection): Randomly select, at each node, F
attributes as candidates for the split at the node. The CART
methodology is used to grow the trees to maximum size
■ Forest-RC (random linear combinations): Creates new attributes (or
features) that are a linear combination of the existing attributes
(reduces the correlation between individual classifiers)
■ Comparable in accuracy to Adaboost, but more robust to errors and outliers
■ Insensitive to the number of attributes selected for consideration at each
split, and faster than bagging or boosting
116
 Classification of Class-Imbalanced Data Sets
■ Class-imbalance problem: Rare positive example but numerous
negative ones, e.g., medical diagnosis, fraud, oil-spill, fault, etc.
■ Traditional methods assume a balanced distribution of classes
and equal error costs: not suitable for class-imbalanced data
■ Typical methods for imbalance data in 2-class classification:
■ Oversampling: re-sampling of data from positive class

■ Under-sampling: randomly eliminate tuples from negative

class
■ Threshold-moving: moves the decision threshold, t, so that
the rare class tuples are easier to classify, and hence, less
chance of costly false negative errors
■ Ensemble techniques: Ensemble multiple classifiers
introduced above
■ Still difficult for class imbalance problem on multiclass tasks
117
 Chapter 8. Classification: Basic Concepts

■ Classification: Basic Concepts

■ Decision Tree Induction
■ Bayes Classification Methods
■ Rule-Based Classification
■ Model Evaluation and Selection
■ Techniques to Improve Classification Accuracy:
Ensemble Methods
■ Summary
118
 Summary (I)
■ Classification is a form of data analysis that extracts models
describing important data classes.
■ Effective and scalable methods have been developed for decision
tree induction, Naive Bayesian classification, rule-based
classification, and many other classification methods.
■ Evaluation metrics include: accuracy, sensitivity, specificity,
precision, recall, F measure, and Fß measure.
■ Stratified k-fold cross-validation is recommended for accuracy
estimation. Bagging and boosting can be used to increase overall
accuracy by learning and combining a series of individual models.

119
 Summary (II)
■ Significance tests and ROC curves are useful for model selection.
■ There have been numerous comparisons of the different
classification methods; the matter remains a research topic
■ No single method has been found to be superior over all others
for all data sets
■ Issues such as accuracy, training time, robustness, scalability,
and interpretability must be considered and can involve trade-
offs, further complicating the quest for an overall superior
method

120
 Reference: Books on Classification
■ E. Alpaydin. Introduction to Machine Learning, 2 nd ed., MIT Press, 2011
■ L. Breiman, J. Friedman, R. Olshen, and C. Stone. Classification and Regression Trees.
Wadsworth International Group, 1984.
■ C. M. Bishop. Pattern Recognition and Machine Learning. Springer, 2006.
■ R. O. Duda, P. E. Hart, and D. G. Stork. Pattern Classification, 2ed. John Wiley, 2001
■ T. Hastie, R. Tibshirani, and J. Friedman. The Elements of Statistical Learning: Data Mining,
Inference, and Prediction. Springer-Verlag, 2001
■ H. Liu and H. Motoda (eds.). Feature Extraction, Construction, and Selection: A Data Mining
Perspective. Kluwer Academic, 1998T. M. Mitchell. Machine Learning. McGraw Hill, 1997
■ S. Marsland. Machine Learning: An Algorithmic Perspective. Chapman and Hall/CRC, 2009.
■ J. R. Quinlan. C4.5: Programs for Machine Learning. Morgan Kaufmann, 1993
■ J. W. Shavlik and T. G. Dietterich. Readings in Machine Learning. Morgan Kaufmann, 1990.
■ P. Tan, M. Steinbach, and V. Kumar. Introduction to Data Mining. Addison Wesley, 2005.
■ S. M. Weiss and C. A. Kulikowski. Computer Systems that Learn: Classification and Prediction
Methods from Statistics, Neural Nets, Machine Learning, and Expert Systems. Morgan
Kaufman, 1991.
■ S. M. Weiss and N. Indurkhya. Predictive Data Mining. Morgan Kaufmann, 1997.
■ I. H. Witten and E. Frank. Data Mining: Practical Machine Learning Tools and Techniques, 2ed.
Morgan Kaufmann, 2005.
121
 Reference: Decision-Trees
■ M. Ankerst, C. Elsen, M. Ester, and H.-P. Kriegel. Visual classification: An interactive approach to decision tree
construction. KDD'99
■ C. Apte and S. Weiss. Data mining with decision trees and decision rules. Future Generation Computer Systems,
13, 1997
■ C. E. Brodley and P. E. Utgoff. Multivariate decision trees. Machine Learning, 19:45–77, 1995.
■ P. K. Chan and S. J. Stolfo. Learning arbiter and combiner trees from partitioned data for scaling machine
learning. KDD'95
■ U. M. Fayyad. Branching on attribute values in decision tree generation. AAAI’94
■ M. Mehta, R. Agrawal, and J. Rissanen. SLIQ : A fast scalable classifier for data mining. EDBT'96.
■ J. Gehrke, R. Ramakrishnan, and V. Ganti. Rainforest: A framework for fast decision tree construction of large
datasets. VLDB’98.
■ J. Gehrke, V. Gant, R. Ramakrishnan, and W.-Y. Loh, BOAT -- Optimistic Decision Tree Construction. SIGMOD'99.
■ S. K. Murthy, Automatic Construction of Decision Trees from Data: A Multi-Disciplinary Survey, Data Mining and
Knowledge Discovery 2(4): 345-389, 1998
■ J. R. Quinlan. Induction of decision trees. Machine Learning, 1:81-106, 1986
■ J. R. Quinlan and R. L. Rivest. Inferring decision trees using the minimum description length principle.
Information and Computation, 80:227–248, Mar. 1989
■ S. K. Murthy. Automatic construction of decision trees from data: A multi-disciplinary survey. Data Mining and
Knowledge Discovery, 2:345–389, 1998.
■ R. Rastogi and K. Shim. Public: A decision tree classifier that integrates building and pruning. VLDB’98.
■ J. Shafer, R. Agrawal, and M. Mehta. SPRINT : A scalable parallel classifier for data mining. VLDB’96
■ Y.-S. Shih. Families of splitting criteria for classification trees. Statistics and Computing, 9:309–315, 1999.
122
 Reference: Neural Networks

■ C. M. Bishop, Neural Networks for Pattern Recognition. Oxford University

Press, 1995
■ Y. Chauvin and D. Rumelhart. Backpropagation: Theory, Architectures, and
Applications. Lawrence Erlbaum, 1995
■ J. W. Shavlik, R. J. Mooney, and G. G. Towell. Symbolic and neural learning
algorithms: An experimental comparison. Machine Learning, 6:111–144,
1991
■ S. Haykin. Neural Networks and Learning Machines. Prentice Hall, Saddle
River, NJ, 2008
■ J. Hertz, A. Krogh, and R. G. Palmer. Introduction to the Theory of Neural
Computation. Addison Wesley, 1991.
■ R. Hecht-Nielsen. Neurocomputing. Addison Wesley, 1990
■ B. D. Ripley. Pattern Recognition and Neural Networks. Cambridge
University Press, 1996

123
 Reference: Support Vector Machines

■ C. J. C. Burges. A Tutorial on Support Vector Machines for Pattern

Recognition. Data Mining and Knowledge Discovery, 2(2): 121-168, 1998
■ N. Cristianini and J. Shawe-Taylor. An Introduction to Support Vector
Machines and Other Kernel-Based Learning Methods. Cambridge Univ. Press,
2000.
■ H. Drucker, C. J. C. Burges, L. Kaufman, A. Smola, and V. N. Vapnik. Support
vector regression machines, NIPS, 1997
■ J. C. Platt. Fast training of support vector machines using sequential minimal
optimization. In B. Schoelkopf, C. J. C. Burges, and A. Smola, editors,
Advances in Kernel Methods|Support Vector Learning, pages 185–208. MIT
Press, 1998
■ B. Schl¨okopf, P. L. Bartlett, A. Smola, and R. Williamson. Shrinking the tube:
A new support vector regression algorithm. NIPS, 1999.
■ H. Yu, J. Yang, and J. Han. Classifying large data sets using SVM with
hierarchical clusters. KDD'03.

124
 Reference: Pattern-Based
Classification
■ H. Cheng, X. Yan, J. Han, and C.-W. Hsu, Discriminative Frequent Pattern Analysis
for Effective Classification, ICDE'07
■ H. Cheng, X. Yan, J. Han, and P. S. Yu, Direct Discriminative Pattern Mining for
Effective Classification, ICDE'08
■ G. Cong, K.-L. Tan, A. K. H. Tung, and X. Xu. Mining top-k covering rule groups for
gene expression data. SIGMOD'05
■ G. Dong and J. Li. Efficient mining of emerging patterns: Discovering trends and
differences. KDD'99
■ H. S. Kim, S. Kim, T. Weninger, J. Han, and T. Abdelzaher. NDPMine: Efficiently
mining discriminative numerical features for pattern-based classification.
ECMLPKDD'10
■ W. Li, J. Han, and J. Pei, CMAR: Accurate and Efficient Classification Based on
Multiple Class-Association Rules, ICDM'01
■ B. Liu, W. Hsu, and Y. Ma. Integrating classification and association rule mining.
KDD'98
■ J. Wang and G. Karypis. HARMONY: Efficiently mining the best rules for
classification. SDM'05
125
 References: Rule Induction
■ P. Clark and T. Niblett. The CN2 induction algorithm. Machine Learning, 3:261–283,
1989.
■ W. Cohen. Fast effective rule induction. ICML'95
■ S. L. Crawford. Extensions to the CART algorithm. Int. J. Man-Machine Studies,
31:197–217, Aug. 1989
■ J. R. Quinlan and R. M. Cameron-Jones. FOIL: A midterm report. ECML’93
■ P. Smyth and R. M. Goodman. An information theoretic approach to rule induction.
IEEE Trans. Knowledge and Data Engineering, 4:301–316, 1992.
■ X. Yin and J. Han. CPAR: Classification based on predictive association rules. SDM'03

126
 References: K-NN & Case-Based
Reasoning

■ A. Aamodt and E. Plazas. Case-based reasoning: Foundational

issues, methodological variations, and system approaches. AI
Comm., 7:39–52, 1994.
■ T. Cover and P. Hart. Nearest neighbor pattern classification.
IEEE Trans. Information Theory, 13:21–27, 1967
■ B. V. Dasarathy. Nearest Neighbor (NN) Norms: NN Pattern
Classication Techniques. IEEE Computer Society Press, 1991
■ J. L. Kolodner. Case-Based Reasoning. Morgan Kaufmann, 1993
■ A. Veloso, W. Meira, and M. Zaki. Lazy associative classification.
ICDM'06

127
eferences: Bayesian Method & Statistical Model
■ A. J. Dobson. An Introduction to Generalized Linear Models. Chapman & Hall, 1990.
■ D. Heckerman, D. Geiger, and D. M. Chickering. Learning Bayesian networks: The
combination of knowledge and statistical data. Machine Learning, 1995.
■ G. Cooper and E. Herskovits. A Bayesian method for the induction of probabilistic
networks from data. Machine Learning, 9:309–347, 1992
■ A. Darwiche. Bayesian networks. Comm. ACM, 53:80–90, 2010
■ A. P. Dempster, N. M. Laird, and D. B. Rubin. Maximum likelihood from incomplete data
via the EM algorithm. J. Royal Statistical Society, Series B, 39:1–38, 1977
■ D. Heckerman, D. Geiger, and D. M. Chickering. Learning Bayesian networks: The
combination of knowledge and statistical data. Machine Learning, 20:197–243, 1995
■ F. V. Jensen. An Introduction to Bayesian Networks. Springer Verlag, 1996.
■ D. Koller and N. Friedman. Probabilistic Graphical Models: Principles and Techniques.
The MIT Press, 2009
■ J. Pearl. Probabilistic Reasoning in Intelligent Systems. Morgan Kauffman, 1988
■ S. Russell, J. Binder, D. Koller, and K. Kanazawa. Local learning in probabilistic networks
with hidden variables. IJCAI'95
■ V. N. Vapnik. Statistical Learning Theory. John Wiley & Sons, 1998.
128
 Refs: Semi-Supervised & Multi-Class
Learning

■ O. Chapelle, B. Schoelkopf, and A. Zien. Semi-supervised

Learning. MIT Press, 2006
■ T. G. Dietterich and G. Bakiri. Solving multiclass learning
problems via error-correcting output codes. J. Articial
Intelligence Research, 2:263–286, 1995
■ W. Dai, Q. Yang, G. Xue, and Y. Yu. Boosting for transfer
learning. ICML’07
■ S. J. Pan and Q. Yang. A survey on transfer learning. IEEE Trans.
on Knowledge and Data Engineering, 22:1345–1359, 2010
■ B. Settles. Active learning literature survey. In Computer
Sciences Technical Report 1648, Univ. Wisconsin-Madison, 2010
■ X. Zhu. Semi-supervised learning literature survey. CS Tech.
Rep. 1530, Univ. Wisconsin-Madison, 2005
129
Refs: Genetic Algorithms & Rough/Fuzzy
Sets
■ D. Goldberg. Genetic Algorithms in Search, Optimization, and Machine
Learning. Addison-Wesley, 1989
■ S. A. Harp, T. Samad, and A. Guha. Designing application-specific neural
networks using the genetic algorithm. NIPS, 1990
■ Z. Michalewicz. Genetic Algorithms + Data Structures = Evolution Programs.
Springer Verlag, 1992.
■ M. Mitchell. An Introduction to Genetic Algorithms. MIT Press, 1996
■ Z. Pawlak. Rough Sets, Theoretical Aspects of Reasoning about Data. Kluwer
Academic, 1991
■ S. Pal and A. Skowron, editors, Fuzzy Sets, Rough Sets and Decision Making
Processes. New York, 1998
■ R. R. Yager and L. A. Zadeh. Fuzzy Sets, Neural Networks and Soft Computing.
Van Nostrand Reinhold, 1994

130
 References: Model Evaluation, Ensemble
Methods
■ L. Breiman. Bagging predictors. Machine Learning, 24:123–140, 1996.
■ L. Breiman. Random forests. Machine Learning, 45:5–32, 2001.
■ C. Elkan. The foundations of cost-sensitive learning. IJCAI'01
■ B. Efron and R. Tibshirani. An Introduction to the Bootstrap. Chapman & Hall, 1993.
■ J. Friedman and E. P. Bogdan. Predictive learning via rule ensembles. Ann. Applied
Statistics, 2:916–954, 2008.
■ T.-S. Lim, W.-Y. Loh, and Y.-S. Shih. A comparison of prediction accuracy, complexity,
and training time of thirty-three old and new classification algorithms. Machine
Learning, 2000.
■ J. Magidson. The Chaid approach to segmentation modeling: Chi-squared automatic
interaction detection. In R. P. Bagozzi, editor, Advanced Methods of Marketing
Research, Blackwell Business, 1994.
■ J. R. Quinlan. Bagging, boosting, and c4.5. AAAI'96.
■ G. Seni and J. F. Elder. Ensemble Methods in Data Mining: Improving Accuracy
Through Combining Predictions. Morgan and Claypool, 2010.
■ Y. Freund and R. E. Schapire. A decision-theoretic generalization of on-line learning
and an application to boosting. J. Computer and System Sciences, 1997
131
Surplus Slides

132
 Sequential Covering Algorithm

while (enough target tuples left)

generate a rule
remove positive target tuples satisfying this rule

Examples
Examples covered
covered by Rule 2
Examples
by Rule 1 covered
by Rule 3
Positive
examples

133
 Rule Generation
■ To generate a rule
while(true)
find the best predicate p
if foil-gain(p) > threshold then add p to current rule
else break

A3=1&&A
1=2
A3=1&&A1=2
&&A8=5A3=1

Positive Negative
examples examples

134
 How to Learn-One-Rule?
■ Start with the most general rule possible: condition = empty
■ Adding new attributes by adopting a greedy depth-first strategy
■ Picks the one that most improves the rule quality

■ Rule-Quality measures: consider both coverage and accuracy

■ Foil-gain (in FOIL & RIPPER): assesses info_gain by extending

condition

■ favors rules that have high accuracy and cover many positive tuples
■ Rule pruning based on an independent set of test tuples

Pos/neg are # of positive/negative tuples covered by R.

If FOIL_Prune is higher for the pruned version of R, prune R
135
 Issues: Evaluating Classification Methods

■ Accuracy
■ classifier accuracy: predicting class label

■ predictor accuracy: guessing value of predicted attributes

■ Speed
■ time to construct the model (training time)

■ time to use the model (classification/prediction time)

■ Robustness: handling noise and missing values

■ Scalability: efficiency in disk-resident databases
■ Interpretability
■ understanding and insight provided by the model

■ Other measures, e.g., goodness of rules, such as decision tree

size or compactness of classification rules

136
 Gain Ratio for Attribute Selection (C4.5)
(MK:contains errors)

■ Information gain measure is biased towards attributes with a

large number of values
■ C4.5 (a successor of ID3) uses gain ratio to overcome the
problem (normalization to information gain)

■ GainRatio(A) = Gain(A)/SplitInfo(A)
■ Ex.
■ gain_ratio(income) = 0.029/0.926 = 0.031
■ The attribute with the maximum gain ratio is selected as the
splitting attribute
137
 Gini index (CART, IBM IntelligentMiner)

■ Ex. D has 9 tuples in buys_computer = “yes” and 5 in “no”

■ Suppose the attribute income partitions D into 10 in D 1 : {low, medium} and 4

in D 2

but gini{medium,high} is 0.30 and thus the best since it is the lowest
■ All attributes are assumed continuous-valued
■ May need other tools, e.g., clustering, to get the possible split values
■ Can be modified for categorical attributes

138
 Predictor Error Measures

■ Measure predictor accuracy: measure how far off the predicted value is from
the actual known value
■ Loss function: measures the error betw. y i and the predicted value y i’
■ Absolute error: | y i – yi’|
■ Squared error: (y i – yi’)2
■ Test error (generalization error): the average loss over the test set
■ Mean absolute error: Mean squared error:

■ Relative absolute error: Relative squared error:

The mean squared-error exaggerates the presence of outliers

Popularly use (square) root mean-square error, similarly, root relative
squared error

139
 Scalable Decision Tree Induction
Methods

■ SLIQ (EDBT’96 — Mehta et al.)

■ Builds an index for each attribute and only class list and the

current attribute list reside in memory

■ SPRINT (VLDB’96 — J. Shafer et al.)
■ Constructs an attribute list data structure

■ PUBLIC (VLDB’98 — Rastogi & Shim)

■ Integrates tree splitting and tree pruning: stop growing the

tree earlier
■ RainForest (VLDB’98 — Gehrke, Ramakrishnan & Ganti)
■ Builds an AVC-list (attribute, value, class label)

■ BOAT (PODS’99 — Gehrke, Ganti, Ramakrishnan & Loh)

■ Uses bootstrapping to create several small samples

140
 Data Cube-Based Decision-Tree
Induction
■ Integration of generalization with decision-tree induction
(Kamber et al.’97)
■ Classification at primitive concept levels
■ E.g., precise temperature, humidity, outlook, etc.
■ Low-level concepts, scattered classes, bushy classification-
trees
■ Semantic interpretation problems
■ Cube-based multi-level classification
■ Relevance analysis at multi-levels
■ Information-gain analysis with dimension + level

141