Abstract
The similarity computation has an important influence on the performance of the personalized recommendation system. Due to the sparsity of the data in the recommendation system, the traditional similarity measurement method can not achieve the desired results. Sparse scoring similarity measurement method based on the Bhattacharyya measure has achieved good results. In this paper, the score similarity measure is introduced. Then, a method to improve the similarity measure based on the Bhattacharyya measure is presented. Taking into account all the items of the two users’ score, and according to the score to distinguish between the user is really interested in the project.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have