Applying evolutionary algorithms to the problem of information filtering

Author

Tjoa, A. Min ; Höfferer, Max ; Ehrentraut, Gunter ; Untersmeyer, Peter

Author_Institution

Inst. of Software Technol., Vienna Univ. of Technol., Austria

fYear

1997

Firstpage

450

Lastpage

458

Abstract

This paper presents an intelligent information filtering system that learns from user feedback and behavior through evolutionary algorithms. By applying the learning abilities of a classifier system and genetic algorithms to the system, the following tasks can be performed: (1) reducing a user´s information overload; (2) predicting the actions that the users are supposed to do; and (3) prioritizing electronic mail.

Keywords

cognitive systems; data reduction; electronic mail; feedback; genetic algorithms; learning (artificial intelligence); online front-ends; pattern classification; user modelling; CIFS; Cognitive Information Filtering System; classifier system; electronic mail prioritization; evolutionary algorithms; genetic algorithms; intelligent information filtering system; learning; user action prediction; user behavior; user feedback; user information overload reduction; Computer science; Evolutionary computation; Expert systems; Feedback; Genetic algorithms; Indexing; Information filtering; Information filters; Information retrieval; Intelligent systems;

fLanguage

English

Publisher

ieee

Conference_Titel

Database and Expert Systems Applications, 1997. Proceedings., Eighth International Workshop on

Conference_Location

Toulouse, France

Print_ISBN

0-8186-8147-0

Type

conf

DOI

10.1109/DEXA.1997.617331

Filename

617331