Social Recommendation Based on Quantified Trust and User’s Primary Preference Space

Social recommendation has received great attention recently, which uses social information to alleviate the data sparsity problem and cold-start of systems. However, existing methods have two deficiencies. First, binary trust network used by current cannot reflect level different users. Second, assume that users only consider same influencial factors when purchasing goods establishing friendships, does not match reality, since may preferences in scenarios. To address these issues, this paper, we propose a novel framework based on preference, named TPSR, including quantify method random walk with restart (TQ_RWR) user’s primary preference space model (UPPS). Our experimental results four public real-world datasets show TQ_RWR can improve utilization information, recommended accuracy. In addition, compared methods/studies, TPSR achieve higher performance metrics, root mean square error, precision, recall F1 value.