Using Topic Keyword Clusters for Automatic Document Clustering
ICITA '05 Proceedings of the Third International Conference on Information Technology and Applications (ICITA'05) Volume 2 - Volume 02
Genetic Algorithms and Genetic Programming: Modern Concepts and Practical Applications
Genetic Algorithms and Genetic Programming: Modern Concepts and Practical Applications
On the performance of evolutionary algorithms in biomedical keyword clustering
Proceedings of the 13th annual conference companion on Genetic and evolutionary computation
A fast and elitist multiobjective genetic algorithm: NSGA-II
IEEE Transactions on Evolutionary Computation
| Hi-index | 0.00 |
As it is not trivial to cope with the fast growing number of papers published in the field of medicine and biology, intelligent search strategies are needed to be able to access the required information as fast and accurately as possible. In [5] we have proposed a method for keyword clustering as a first step towards an intelligent search strategy in biomedical information retrieval. In this paper we focus on the analysis of the internal dynamics of the evolutionary algorithms applied here using solution encoding specific population diversity analysis, which is also defined in this paper. The population diversity results obtained using evolution strategies, genetic algorithms, genetic algorithms with offspring selection and also a multi-objective approach, the NSGA-II, are discussed here. We see that the diversity of the populations is preserved over the generations, decreasing towards the end of the runs, which indicates a good performance of the selection process.