一种融合项目信息与信任机制的协同过滤算法

A collaborative filtering algorithm combining project in-formation and trust mechanism

针对现有方法未能考虑用户社会地位和信任对象的差异及用户相似性在面对不同项目时不能自适应变化的问题,提出一种融合项目信息与信任机制的协同过滤算法CF-PIC。首先,将项目按照所属领域进行划分,综合考虑用户在不同领域内的全局信任度和局部信任度,构建特定领域的信任网络;然后,将项目间相似性融入Pearson相关系数,计算用户面对不同项目时的偏好程度,以更加精确地捕获其近邻信息;最后,对目标用户进行TOP-N推荐。在真实数据集Epinions上的大量实验表明,该算法的推荐性能相较于经典的协同过滤算法和融入单一信息的算法有了大幅提高。

The existing methods fail to take into account the differences in the user's social status and trust objects,and the problem that the user's similarity cannot adapt to changes in the face of different projects.To address the issues,a collaborative filtering algorithm combining project information and trust mechanism(CF-PIC)is proposed.Firstly,items are divided according to their fields.Trust network of the specific field is built by considering the global trust and local trust of users in different fields.Then,in order to capture the neighbor information more accurately,the similarity be⁃tween items is integrated into Pearson correlation coefficient to calculate the degree of user's preference when facing dif⁃ferent items.Finally,TOP-N recommendation is made to the target user.Experimental results using the Epinions dataset show that the performance of the proposed algorithm is greatly improved,compared to the classical collaborative filtering algorithm and the algorithm incorporating single information.