基于聚类和奖惩用户模型的协同过滤算法

Collaborative Filtering Algorithm Based on Clustering and Incentive/Penalty User Model

根据用户体验为其推荐感兴趣的项目是推荐系统中最重要的问题.本文提出了一种新的易于实现的CBCF(Clustering-Based CF)算法,该算法基于激励/惩罚用户(IPU)模型进行推荐.本文旨在通过IPU模型深入研究用户间偏好的差异来提高准确率、召回率和F1-score方面的性能.本文提出了一个约束优化问题,目标是在给定的精度下最大限度地提高召回率(或F1-score).为此,根据实际评分数据和皮尔逊相关系数,将用户分为若干用户簇,然后根据同一用户簇的偏好倾向,对每个项目进行奖励/处罚.实验结果表明,本文提出的算法在给定准确率的条件下,召回率可以显著提高50%左右.

Giving or recommending appropriate content based on the quality of experience is the most important in recommender systems.This study proposes a new CBCF(Clustering-Based CF)method using an Incentivized/Penalized User(IPU)model,which is thus easy to implement.The purpose of this study is to improve recommendation performance of accuracy,recall and F1-score by studying the differences of users’preferences through IPU model.This study formulates a constrained optimization problem in which we aim to maximize the recall(or equivalently F1-score)for a given precision.To this end,users are divided into several clusters based on the actual rating data and Pearson correlation coefficient.Afterward,we give each item an incentive/penalty according to the preference tendency by users within the same cluster.Experiments show that under the condition of given accuracy,the recall rate of the proposed algorithm can be improved by about 50%.