An intelligent system for personalized advertising on the internet

Abstract

An Internet advertising system proposed here clusters Web site users with similar preferences into numerous segments through Web usage mining. It utilizes fuzzy rules which express user segments' surfing patterns and appropriate Web ads. It selects proper Web ads by fuzzy inference, stores them in recommendation sets database, and forwards them to the target user. To verify the effectiveness of the system, changes in click-through-ratio scores per e-newspaper section are observed. © Springer-Verlag 2004.