Publication details

Searching for Significant Word Associations in Text Documents Using Genetic Algorithms

Authors

ŽIŽKA Jan ŠRÉDL Michal BOUREK Aleš

Year of publication 2003
Type Article in Proceedings
Conference Computional Linguistics and Intelligent Text Processing
MU Faculty or unit

Faculty of Informatics

Citation
Field Informatics
Keywords machine learning; text document processing; genetic algorithms; naive Bayes method
Description The paper describes experiments that used Genetic Algorithms for looking for important word assocoations (phrases) in unstructured text documents obtained from the Internet in the area of a specialized medicine branch. Genetic alforithms can evolve sets of word associations with assigned significance weights from the document categorization point of view (relevant and irrelevant documents). The categorization is similarly reliable like the naive Bayes classification based on individual words. In addition, genetic algorithms provided phrases consisting of one, two, and three words. The phrases were quite meaningful from the human point of view.
Related projects:

You are running an old browser version. We recommend updating your browser to its latest version.

More info