GENETIC ALGORITHM PARAMETER SETS FOR LINE LABELING

This paper concerns the use of genetic algorithms for line labelling. We are interested in finding an optimal set of algorithm control param eters for this problem. We give results from using a simple genetic al gorithm to solve several line labelling problems and discuss the effec ts of crossover type, population size, crossover rate, mutation rate a nd iteration limit on algorithm performance. We conclude that the algo rithm is very sensitive to mutation rate, and that there is a threshol d population size beyond which success rates are very high but that th is threshold increases rapidly with the problem size. We recommend tha t a mutation rate of 0.02 be used in conjunction with a crossover rate of between 0.6 and 0.9. Iteration limit should initially be high, and should only be lowered when the other parameters have been tuned. (C) 1997 Elsevier Science B.V.