基于SVM的项目协同过滤算法

iming at the characteristics of the Internet reservation system and the existing e-commerce system users and items sharp growth in the number of data sparsity problem, proposed a item-based collaborative filtering algorithm based on SVM. In this method, the SVM algorithm is used in item-based collaborative filtering recommendation algorithm, SVM trains model by using existing data, then use the trained model to calculate the similarity between items, and predict the scores of users on the item which was not scoring, resulting recommendations. Simulation experiment illustrates the method has a lower mean absolute error and a higher precisions, indicates the validity of the method. And the method is applied to the online reservation system.)