Rating information entropy for cold-start recommendation

To solve the cold-start problem in collaborative filtering recommender system and the unilateral problem in existing method for cold-start recommendation, we proposed a rating information entropy method. In rating matrix, we use the theory of information entropy to calculate the information entropy of both user level and item level. These two factors are integrated to get the rating information entropy that is used as the weight of rating. Experimental results show that the proposed method can solve the unilateral problem in existing entropy method and improve the accuracy and rationality of the recommendation for the new user. © 2011 Binary Information Press.