|国家预印本平台
首页|一种基于时间效应和用户兴趣变化的改进推荐算法

一种基于时间效应和用户兴趣变化的改进推荐算法

Recommended Algorithm based on Time Effect and Changes in user interest

中文摘要英文摘要

该文针对传统的协同过滤推荐算法较少考虑用户兴趣的变化而导致推荐结果不理想的问题,提出一种基于时间效应和用户兴趣变化的改进推荐算法。该算法通过借鉴心理学上的遗忘规律,依据评价时间确定每项评分的重要性,将用户的兴趣变化用遗忘规律进行模拟,并与协同过滤算法进行结合,综合考虑时间效应和用户的兴趣变化对用户之间相似度的影响,最终为用户提供高效的推荐。实验结果表明,与传统的协同过滤推荐算法相比,该算法提高了推荐结果的准确性。

ccording to the problem of less consideration changes in user interests in the traditional collaborative filtering recommendation algorithm, a recommended algorithm based on time effect and changes in user interest is proposed. It simulated changes in user interests based on the forgotten laws on psychology, and with the combination of collaborative filtering algorithms, considering the effect of similarity between the users under the time effect and the interest change, providing efficient recommendation for users at last.Experimental results show that the algorithm based on time effect and changes in user interest improves the recommendation accuracy Compared with the traditional collaborative filtering algorithm.

孙光辉、孟祥武

计算技术、计算机技术

推荐系统协同过滤兴趣变化时间效应遗忘曲线

Recommended systemsCollaborative filteringTime effectChanges in interestForgetting Curve

孙光辉,孟祥武.一种基于时间效应和用户兴趣变化的改进推荐算法[EB/OL].(2013-12-23)[2025-08-28].http://www.paper.edu.cn/releasepaper/content/201312-681.点此复制

评论