Intelligent information filtering via hybrid techniques: Hill climbing, case-based reasoning, index patterns, and genetic algorithms

As the size of the Internet increases, the amount of data available to users has dramatically risen, resulting in an information overload for users. This work shows that information overload is a problem, and that data is organized poorly by existing browsers. To address these problems, an intelligent information news filtering system named INFOS (Intelligent News Filtering Organizational System) was created to reduce the user's search burden by automatically eliminating Usenet news articles predicted to be irrelevant. These predictions are learned automatically by adapting an internal user model that is based upon features taken from articles and collaborative features derived from other users. The features are manipulated through keyword-based techniques, knowledge-based techniques, and genetic algorithms to build a user model to perform the actual filtering. The integration of knowledge-based techniques for in-depth analysis, statistical and keyword approaches for scalability, and genetic algorithms for exploration allows INFOS to achieve better filtering performance than by using either technique alone. Experimental results collected from the prototype of INFOS validate the gain in performance within the domain of news articles posted to electronic bulletin boards.