«上一篇/Previous Article|本期目录/Table of Contents|下一篇/Next Article»

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:

24

期数:

2014年01期

页码:

35-38

栏目:

智能、算法、系统工程

出版日期:

2014-01-31

文章信息/Info

Title:

Recommendation Algorithm of Combining Clustering with Collaborative Filtering Based on Genetic Algorithm

文章编号:

1673-629X（2014）01-0035-04

作者:

冯智明; 苏一丹; 覃华; 邓海

广西大学 计算机与电子信息学院

Author(s):

FENG Zhi-ming; SU Yi-dan; QIN Hua; DENG Hai

关键词:

遗传算法; k均值聚类; item-based协同过滤; 项目推荐

Keywords:

genetic algorithm; k-means clustering; item-based collaborative filtering; item recommendation

分类号:

TP301.6

文献标志码:

A

摘要:

使用协同过滤进行推荐，在处理大数据集时存在效率问题和推荐结果质量不高的问题。 k均值聚类在处理大数据集时有着较好的性能。针对使用协同过滤进行推荐存在的问题，通过使用遗传算法将聚类和协同过滤组合起来进行项目推荐，以此来提高推荐算法的推荐效率和推荐质量，降低组合聚类和协同过滤进行推荐的复杂度。使用组合得到的算法在MovieLens数据集上做推荐对比实验，结果表明，相比单纯使用协同过滤进行推荐，使用基于遗传算法的聚类与协同过滤组合推荐算法进行项目推荐，能得到质量更好的推荐结果。

Abstract:

When dealing with item recommendation with large data sets,there are problems of efficiency and the low quality of the results for collaborative filtering. K-means clustering has a better performance when processing large data sets. In order to solve problems of col-laborative filtering,genetic algorithm can be used to combine clustering and collaborative filtering for item recommendation to improve the efficiency and quality of the recommendation algorithm,reduce the complexity of item recommendation by the combination of cluste-ring and collaborative filtering. Do comparative experiments using the combination algorithm in Movielens data sets. The experimental re-sults show that,compared with pure collaborative filtering recommendation,using genetic algorithm to combine clustering with collabora-tive filtering for item recommendation can get a better quality results.

常用功能

更新日期/Last Update: 1900-01-01