SuJeong Ko, Junghyun Lee. Discovery of User Preference through Genetic Algorithm and Bayesian Categorization for Recommendation. In Hiroshi Arisawa, Yahiko Kambayashi, Vijay Kumar, Heinrich C. Mayr, Ingrid Hunt, editors, ER 2001 Workshops, HUMACS, DASWIS, ECOMO, and DAMA, Yokohama Japan, November 27-30, 2001, Revised Papers. Volume 2465 of Lecture Notes in Computer Science, pages 471-484, Springer, 2001. [doi]
Recent recommender system uses two complementary techniques. Collaborative filtering uses a database about user preferences to predict additional topics. Content based systems provide recommendations by matching user interests with topic attributes. In this paper, we describe a method for discovery of user preference by using hybrid two techniques for recommendation that allow the application of machine learning algorithm. The method generates recommendations based on clustering user and categorizing items with feature selection through association word mining by Apriori algorithm. We use Genetic algorithm to group users based on items categorized by Naïve Bayes classifier. Then, we recommend web documents to user based on grouped user preference and information of categorized items. We evaluate our method on a large database of user ratings for web document and it significantly outperforms previous proposed methods.