A fully automated recommender system using collaborative filters
Wadee S. Al halabi, Miroslav Kubát, MOIEZ A. TAPIA · Communications, Internet, and Information Technology · 2007
Recommender systems represent valuable marketing tools in e-commerce. No wonder that this field has enjoyed quite some attention from the scientific community that seeks to develop algorithms that optimize the performance of recommender systems based on the analysis of historical data. In the research reported here, we experimented with a mechanism that combines traditional collaborative filtering approaches with explicit rating. This paper describes the technique and illustrates its behavior in a test-bed we created from real-world data. The results are promising.