Genetic Algorithm
Download as PPTX, PDF0 likes76 views
This document discusses genetic algorithms. It begins with an introduction, explaining that genetic algorithms were first introduced by John Holland in the 1960s and incorporate ideas of natural evolution. It then describes the genetic algorithm process, including initially generating a random population of rules represented as bit strings, assessing the fitness of each rule based on classification accuracy, and creating a new population through crossover and mutation of the fittest rules. Finally, it lists some uses of genetic algorithms, such as for classification, optimization, and evaluating other algorithms.
Genetic Algorithm
- 1. GENETIC ALGORITHM Presented to: Presented by: Prakash Prasad Amarendra K Yadav
- 2. Introduction Genetic Algorithm Process Uses of Genetic Algorithm OUTLINES
- 3. INTRODUCTION Genetic Algorithms (GAs) were first introduced by John H. Holland in the 1960s Incorporate ideas of natural evolution
- 4. GENETIC ALGORITHM Initial population created consisting randomly generated rules Each rule represented by string of bits
- 5. GENETIC ALGORITHM Let samples be described by two Boolean attributes: A1 and A2 C1 and C2 be the class Rule: If A1 and not A2 then C2 can be encoded as bit string “100”; two leftmost bits represents attributes A1 and A2 and rightmost bit represent class If not A1 and not A2 then C1 can be encoded as 001 If an attribute has k values, k>2, k bits may be used to encode attributes values
- 6. GENETIC ALGORITHM Based on notion of survival of fittest, new population is formed to consist of fittest rules in current population, as well as offspring of these rules. Fitness of rule is assesses by its classification accuracy on set of training samples Offspring created by applying genetic operators such as crossover and mutation Crossover substrings from pair of rules are swapped to form new pair rules Mutation ,randomly selected bits in rule string are inverted
- 7. CROSSOVER
- 8. MUTATION
- 9. USES OF GENETIC ALGORITHM Genetic algorithms are easily parallelizable Used for classification Optimization Used to evaluate the fitness of other algorithm
- 10. REFERENCES Data mining. concepts and techniques, 3rd edition(The Morgan Kaufmann Series in Data Management System) https://www.geeksforgeeks.org/genetic-algorithms/ https://www.researchgate.net/figure/Crossover-and-mutation- operations-in-genetic-algorithm_fig2_245282272
- 11. Thank You