基于多属性效用协同过滤系统.docVIP

基于多属性效用协同过滤系统 　　摘要：针对基于多标准的协同过滤（MCCF）推荐系统中用户负担重、超高维问题，提出了基于多属性效用的协同过滤（MAUCF）推荐系统。首先，依据用户浏览行为挖掘属性权重和属性值效用，构造用户的多属性效用函数，获取用户对项目的隐式评分；其次，采用遗传算法（GA）寻找用户偏好的属性值集合；然后，根据属性值集合中属性权重和属性值效用的相似度，寻找最近邻；最后，根据相似度预测最近邻浏览或购买过的项目对目标用户的效用，向目标用户推荐效用大的项目。通过比较实验发现，相对于MCCF，MAUCF挖掘的隐式效用能够替代显式效用，计算维度减少了44.16%，时间消耗减少了27.36%，平均绝对误差（MAE）减少了5.69%，用户满意度提高了13.44%。实验结果表明，MAUCF推荐系统在减少用户负担和计算维度、提高推荐质量方面比MCCF推荐系统表现得更优越。 　　关键词：推荐系统；多属性效用；协同过滤；隐式评分；遗传算法 　　中图分类号： TP391.1 文献标志码：A 　　Abstract： Focusing on user remarkable burden and high dimension of MultiCriteria Collaborative Filtering （MCCF） recommendation system， the recommendation system of MultiAttribute Utility Collaborative Filtering （MAUCF） was proposed. Firstly， attribute weight and attributevalue utility were extracted by user browsing behavior， and users multiattribute utility function was built to achieve implicit rating of items. Secondly， attributevalue collection according to user preference was constructed based on Genetic Algorithm （GA）. Thirdly， the nearest neighborhood was looked for by attribute weight and attributevalue similarity of attributevalue collection. Finally， utilities of items which the nearest neighborhood had browed and bought would be predicted for user by similarity， and the highutility items would be recommended to user. In the comparison experiments with MCCF， the explicit utility was replaced by the implicit utility calculated by MAUCF， calculation dimension decreased by 44.16%， time expense decreased by 27.36%， and Mean Absolute Error （MAE） decreased by 5.69%， and user satisfaction increased by 13.44%. The experimental results show MAUCF recommendation system outperforms MCCF recommendation system on user burden， calculation dimension， and recommendation quality. 　　Key words： recommendation system； multiattribute utility； collaborative filtering； implicit rating； Genetic Algorithm （GA） 　　0 引言 　　随着互联网的快速发展，网络购物成为消费者最重要的购物方式之一。面对购物网站上海量的产品信息，推荐系统帮助用户快速搜寻符合需求的产品。在众多推荐技术中，协同过滤推荐技术的应用最为广泛。而且，基于多标准的协同过滤（MultiCriter