基于用户类别兴趣度的协同过滤推荐算法

摘要

A collaborative filtering recommendation algo-rithm based on users’ category of interest is proposed by the use of educational resources which are classified accord-ing to subject, learner preferences, stability and other char-acteristics. The algorithm firstly calculates the similarity de-gree of user interest categories based on the conditional probability method, and then improve the accuracy of user similarity, and experiments show that the quality of recom-mendation is improved.%对教育资源按照学科进行分类，结合学习者的兴趣偏好集中、稳定等特性，提出一种基于用户类别兴趣度的协同过滤推荐算法。在该算法中，首先根据条件概率的方法计算用户类别兴趣度的相似性，然后将其融合到相关相似性计算公式中，提高用户相似性的准确度，实验表明提高了推荐质量。