Download as pptx, pdf, or txt
Download as pptx, pdf, or txt
You are on page 1of 13

A Genetic Algorithm for

Discovering Classification
rules in Data Mining

1
Genetic Algorithm

2
Genetic Algorithm
Genetic Algorithm (GA) is a search-based
optimization technique based on the principles
of Genetics and Natural Selection.

Terminology
Population:
it is a subset of all the possible solutions to the
given problem.
3
Terminology
Chromosomes
A chromosome is one such solution to the
given problem.

Gene
A chromosome is one such solution to the
given problem.

Allele
It is the value a gene takes for a particular
4
chromosome.
5
Classification Rules
Classification rules are defined to
categorize work into service classes, and
optionally report classes, based on work
qualifiers.
Classification rules are if then rules
The if part states a condition over the data
The then part includes a class label

6
Genetic Operators
Selection
Crossover
Mutation

7
Problem Statement
Finding antecedent part in classification
rules and using that rules in Data Mining
for better results.

8
Approach
Genetic Algorithm generates a population of
individuals, where each individual represents
a classification rule.
Individual consists of conjunction of
conditions composing a given rule
antecedent.

9
Here,

Ai denotes the ith predictor attribute

Opi denotes the operators( =, >, >= )

Valij denotes jth value of the domain of Attri .

10
Calculate
Confidence(conf)

Coverage(cove)

Comprehensibility(comp)

Fitness
11
Mined rules, conf, cove, comp, Fitness
All these values are used in Data
Mining for better results.

12
Thank You
13

You might also like