基于个性化的多样性优化推荐算法

Recommendation Algorithm for Optimizing Diversity Based on Personalization

针对不同人对多样性的偏好不同,提出一种能够在寻优精度和多样性之间权衡的个性化的多样性优化方法.该方法采用一种依据用户历史偏好和项目类别专家评分的推荐技术,生成包含新颖项目和关联项目的多样化的候选推荐列表,然后依据用户多样化偏好程度进行后过滤技术,筛选出最终的多样化推荐列表.最后,本文通过实验结果对比发现,所提出的方法能够有效地提高推荐列表的多样性,而且能够实现个性化的多样化过程.

A recommendation algorithm for optimizing diversity based on personalization,which is capable of controlling the trade-off between accuracy and diversity,is proposed for different people's preference for diversity.The method,utilizing the recommendation technique which relies on historical user preferences and expert evaluation of project categories,generates a diversified candidate recommendation list including new projects and related projects.Then according to user’s diversity preference,the post-filtering approaches are employed to generate the final diversified recommendation list.Finally,the experiments and evaluation show that the proposed method can effectively improve the diversity of recommender lists,and achieve personalized diversification process.