摘要
协同过滤算法是目前应用最成功的推荐技术之一,但传统的协同过滤算法在推荐过程中认为数据是静态的,且各时期的评分对预测所起到的作用是等同的,导致推荐系统的推荐质量下降.针对上述问题,本文提出了一种基于时间权重模糊递增的协同过滤算法.首先,该算法将项目评分赋予时间属性,并利用项目评分的时间属性计算时间窗口的相似性度;其次,利用时间权重对符合模糊递增规律的评分进行预测,同时本文分两个阶段对时间权重进行求解以达到全局最优和局部最优;最后,通过实验仿真分析,该算法的推荐质量较传统的协同过滤算法有显著提高.
Collaborative filtering algorithm is one of the most successful recommendation techniques.However,the existing collaborative filtering algorithms suggest that ratings produced at different times are weighted equally in different historical period,but this can not reflect the changes of item ratings,which will result in a decrease in the recommended quality of the recommendation system.In this paper,a fuzzy increasing-based collaborative filtering algorithm of time weight is proposed to solve the above problems.First,the algorithm assigns time weights to the item ratings,and the similarity of time windows will be calculated through time property.Then,time weight is used for the prediction of item ratings which follow the law of fuzzy increasing.At the same time,the global optimal solutions and partial optimal solutions of time weights are solved through two steps.Finally,experiments on real datasets and synthetic datasets demonstrate that the effectiveness of our approach is provided to validate.
出处
《小型微型计算机系统》
CSCD
北大核心
2018年第2期255-261,共7页
Journal of Chinese Computer Systems
基金
湖南省自然科学基金项目(2015JJ3010)资助
关键词
协同过滤
模糊递增
时间权重
预测
相似性
collaborative filtering fuzzy increasing time weight prediction similarity
