Collaborative filtering recommendation algorithm based on Bayesian theory

Considering that the effectiveness of the collection of neighbors is an important factor to influence the quality of the recommendation in collaborative filtering recommendation algorithm,a collaborative filtering recommendation method based on Bayesian theory was proposed in this paper.It got the value of the users' preference for a certain characteristic by using Bayesian theory.In the calculation of the similarity degree,it considered users' preferences.And then it calculated the collection of the neighbors.The result shows that it can provide better recommendation quality than traditional item-algorithm.