Most of the information in the eld of life sciences is available only in plaintext, leading to an often difficult search for the appropriate information. Especially when performing a PubMed query, not only lots of publications are presented to the user, but also a huge amount of irrelevant references. To solve this kind of problem, biomedical information retrieval provides various approaches, including query extension. Here, user queries are extended with appropriate additional words, selected from previously generated document or keyword clusters. In this master thesis a new approach is presented, where keyword clusters are generated using evolutionary algorithms. It is a collection of three publications on this scope. "Optimization of Keyword Grouping in Biomedical Information Retrieval Using Evolutionary Algorithms, published in "Proceedings of the 22nd European Modeling and Simulation Symposium, EMSS 2010, Morocco" and written by Viktoria Dorfer, Stephan M. Winkler, Thomas Kern, Gerald Petz, and Patrizia Faschang, describes the underlying problem and the principles of the new approach. In "On the Performance of Evolutionary Algorithms in Biomedical Keyword Clustering", published in "Proceedings of the Genetic and Evolutionary Computation Conference, MedGEC 2011, Dublin" and written by Viktoria Dorfer, Stephan M. Winkler, Thomas Kern, Sophie A. Blank, Gerald Petz, and Patrizia Faschang, the method is intensively tested with various evolutionary algorithms and different parameter settings. Finally, a population diversity analysis is performed on the most promising algorithms in "Analysis of Single-Objective and Multi-Objective Evolutionary Algorithms in Keyword Cluster Optimization", published in "Proceedings of the International Conference on Computer Aided Systems Theory, EUROCAST 2011, Las Palmas" and written by Viktoria Dorfer, Stephan M. Winkler, Thomas Kern, Gerald Petz, and Patrizia Faschang.
|Publication status||Published - 2011|