A new user similarity model to improve the accuracy of collaborative filtering

TLDR

Collaborative filtering relies on identifying similar users or items from a rating matrix, but conventional similarity metrics such as cosine, Pearson, and mean squared difference perform poorly in cold‑start scenarios. This study introduces a new user similarity model aimed at enhancing recommendation accuracy when only a few ratings are available. The model incorporates both local rating context and global user preference patterns, and is evaluated on three real datasets against state‑of‑the‑art similarity measures. Results demonstrate that the proposed similarity model outperforms existing methods in recommendation performance.

Abstract

Collaborative filtering has become one of the most used approaches to provide personalized services for users. The key of this approach is to find similar users or items using user-item rating matrix so that the system can show recommendations for users. However, most approaches related to this approach are based on similarity algorithms, such as cosine, Pearson correlation coefficient, and mean squared difference. These methods are not much effective, especially in the cold user conditions. This paper presents a new user similarity model to improve the recommendation performance when only few ratings are available to calculate the similarities for each user. The model not only considers the local context information of user ratings, but also the global preference of user behavior. Experiments on three real data sets are implemented and compared with many state-of-the-art similarity measures. The results show the superiority of the new similarity model in recommended performance.