Combination of Neural and Evolutionary Methods for Data Organization (1998)
AUTHORS:
Bounsaythip Catherine,
Honkela Timo
BOOKTITLE:
Proceedings of FODO'98, The 5th International Conference on Foundations of Data Organization
PAGES:
20-25
@inproceedings{ Bounsaythip1998, editor = "Tanaka, Katsumi and Ghandeharizadeh, Shahram", author = "Bounsaythip, Catherine and Honkela, Timo", booktitle = "Proceedings of FODO'98, The 5th International Conference on Foundations of Data Organization", year = "1998", pages = "20-25", abstract = "Fuzzy logic, articial neural network models and evolutionary computing are the main methodolog- ical tools of the soft computing area. This article provides an overview on two of them, namely neu- ral networks and evolutionary models. The largest number of applications that combine these two is based on the idea that a genetic algorithm is used to optimize the functioning of a neural network. A more general focus is taken here and examples of other kinds of combinations are given. Further- more, special emphasis is given on the combination of genetic algorithms and Kohonen's self-organizing map (SOM). The usage of the SOM in analyzing and organizing the populations produced by a GA is considered in some more detail. Finally, some applications in the Web and information retrieval domain are presented.", title = "Combination of Neural and Evolutionary Methods for Data Organization" }