11. Explain the crossover operation in GA and discuss its importance in genetic modelling.
Crossover is a genetic operator used to combine the genetic information of two parent
chromosomes to produce new offspring. It mimics the process of biological reproduction, where
genes from both parents are inherited by the child.
Importance:
- Encourages exploration of new regions in the search space.
- Helps in combining good traits of two fit individuals.
- Increases diversity while maintaining useful information.
- Generates better-performing offspring.
Types of Crossover:
- Single-Point: Swap tails after one crossover point.
- Two-Point: Swap middle portion between two points.
- N-Point: General form using multiple points.
- Uniform: Randomly select genes from either parent.
12. Describe mutation in genetic algorithms and explain why it is necessary.
Mutation introduces random alterations to individual genes in a chromosome. It is similar to
mutations in nature that result in new traits.
Necessity:
- Prevents premature convergence.
- Maintains genetic diversity.
- Escapes local optima.
- Supports exploration of the search space.
Techniques:
- Flipping (0↔1)
- Reversing a segment
- Interchanging two gene positions.
13. What is the role of population initialization in genetic algorithms?
Population initialization creates the first generation of chromosomes. Usually done randomly, it
ensures diversity.
Importance:
- Wide coverage of the solution space.
- Affects convergence speed and final solution quality.
- Poor initialization may lead to stagnation.
14. Discuss the stopping criteria used in genetic algorithms.
GAs stop when a defined condition is satisfied.
Common Stopping Criteria:
- Maximum number of generations.
- Acceptable solution (fitness) found.
- No significant improvement.
- Time/resource constraints.
15. Compare between Natural and GA terminologies.
Natural Term → GA Term:
- Chromosome → String (solution)
- Gene → Feature/characteristic
- Allele → Value of a gene
- Locus → Position in chromosome
- Genotype → Structure of solution
- Phenotype → Decoded (real) solution
16. Briefly explain the elements of GA.
Elements of GA:
- Encoding: Representation of solutions.
- Initial Population: First generation.
- Fitness Function: Evaluates solution quality.
- Selection: Chooses best individuals.
- Crossover: Combines two parents.
- Mutation: Random alterations.
- Replacement: Forms new population.
17. Explain the Genetic Algorithm Evaluation flow with a diagram.
Evaluation Flow:
1. Start: Random population.
2. Evaluate Fitness.
3. Selection.
4. Crossover.
5. Mutation.
6. Replacement.
7. Test stopping condition.
Loop continues until criteria are met.
18. Explain Binary encoding and Permutation encoding with examples.
Binary Encoding:
- Represents solutions as strings of 1s and 0s.
Example: 10110101
Permutation Encoding:
- Represents ordering problems.
Example: 1 5 3 2 4
19. Briefly discuss Roulette Wheel Selection Algorithm.
Roulette Wheel Selection assigns probability to each chromosome proportional to its fitness.
Steps:
- Calculate total fitness.
- Determine % share of each chromosome.
- Spin the wheel to select.
Higher fitness → More chances of selection.
20. Explain briefly Elitism in GA.
Elitism preserves the best solutions across generations.
Importance:
- Prevents loss of best individuals.
- Ensures performance doesn't degrade.
- Improves convergence speed.
Elite individuals are directly copied to next generation.
