电子学报 ›› 2017, Vol. 45 ›› Issue (12): 3070-3075.DOI: 10.3969/j.issn.0372-2112.2017.12.033

• 科研通信 • 上一篇    下一篇

基于评分预测与排序预测的协同过滤推荐算法

李改1, 陈强3, 李磊2   

  1. 1. 顺德职业技术学院电子与信息工程学院, 广东顺德 528333;
    2. 中山大学数据科学与计算机学院, 广东广州 510006;
    3. 广东第二师范学院计算机科学系, 广东广州 510303
  • 收稿日期:2016-09-13 修回日期:2017-02-28 出版日期:2017-12-25
    • 作者简介:
    • 李改,男,1981年生于湖北松滋.现为顺德职业技术学院电子与信息工程学院副教授.主要研究方向为数据挖掘,推荐系统,机器学习.E-mail:ligai999@126.com;陈强,男,1963年生于广东化州.现为广东第二师范学院计算机系教授,主要研究领域为数据挖掘,知识表示.E-mail:21115903@qq.com
    • 基金资助:
    • 国家自然科学基金 (No.61370186,No.61640222); 广东省自然科学基金项目 (No.2016A030310018); 广东省科技计划项目 (No.2014A010103040,No.2014B010116001); 广州市科技计划项目 (No.201604010049,No.201510010203); 广东第二师范学院教授博士科研专项 (No.2015ARF25)

Collaborative Filtering Recommendation Algorithm Based on Rating Prediction and Ranking Prediction

LI Gai1, CHEN Qiang3, LI Lei2   

  1. 1. Department of Electronic and Information Engineering, Shunde Polytechnic, Foshan, Guangdong 528300, China;
    2. School of Data and Computer Science, Sun Yat-Sen University, Guangzhou, Guangdong 510006, China;
    3. Department of Computer Science, Guangdong University of Education, Guangzhou, Guangdong 510303, China
  • Received:2016-09-13 Revised:2017-02-28 Online:2017-12-25 Published:2017-12-25
    • Supported by:
    • National Natural Science Foundation of China (No.61370186, No.61640222); Program of National Natural Science Foundation of Guangdong Province,  China (No.2016A030310018); Science and Technology Project of Guangdong Province (No.2014A010103040, No.2014B010116001); Guangzhou Science and Technology Plan Project (No.201604010049, No.201510010203); Science Research Program of Professors and Ph.D of Guangdong University of Education (No.2015ARF25)

摘要: 协同过滤推荐算法在电子商务领域运用广泛.之前的研究要么仅从评分预测的角度来研究,要么仅从排序预测的角度来研究.为了兼顾这两个方面,本文在传统的基于评分预测的PMF(Probabilistic Matrix Factorization)算法和基于排序预测的xCLiMF(Extended Collaborative Less-is-More Filtering)算法的基础上提出了一种基于评分预测与排序预测的协同过滤推荐算法URA(Unified Recommendation Algorithm),该方法通过在PMF和xCLiMF算法中共享用户和推荐对象的特征空间,利用PMF算法来学习高精度的用户和推荐对象的特征向量,从而进一步增强排序推荐性能.实验验证,该方法在评价指标NDCG和ERR下均优于PMF和xCLiMF算法,且复杂度与评分点个数线性相关.URA算法可运用于互联网信息推荐领域的大数据处理.

关键词: 推荐系统, 协同排序, 协同过滤, 评分预测, 排序预测

Abstract: Collaborative filtering (CF) recommendation algorithm is widely used in the field of e-commerce.The previous researches on CF focused on either rating prediction or ranking prediction.In order to take into account these two aspects,a collaborative filtering recommendation algorithm based on rating prediction and ranking prediction (Unified Recommendation Algorithm,URA) is proposed.URA shares common latent features of users and items in PMF (Probabilistic Matrix Factorization,rating-oriented) and xCLiMF (Extended Collaborative Less-is-More Filtering,ranking-oriented) algorithms,and PMF learns improved latent features of users and items in URA,so that URA improves the performance of ranking recommendation.Experimental results showed that our proposed URA Algorithm outperformed PMF and xCLiMF algorithms over evaluation metrics NDCG and ERR,and that the complexity of URA is shown to be linear with the number of observed ratings.URA is suitable for big data processing in the field of internet information recommendation.

Key words: recommended systems, collaborative ranking, collaborative filtering, rating prediction, ranking prediction

中图分类号: