User-Item Missing Ratings Complement Based on Two-Dimensional Normal Distribution

The User-Item missing rating data are a kind of uncertain data in e-commerce website, but in recommendation system these missing ratings are the important information when implementing personalized recommendations. Currently, the existing methods are using a fixed value, the average value of all ratings or a predicted value to replace the missing values. In this paper, to solve the issue which considers the ratings factors is unilateral in the existing methods, the missing User-Item rating complement model based on the two-dimensional random variable which is two-dimensional normal distribution is proposed, and the two-dimensional User-Item rating complement algorithm is designed. The experimental results show that this method could effectively resolve low efficiency recommendation caused by the missing User-Item ratings and improve the quality of recommendation significantly in E-commerce recommendation system.