[1]孔元元,白智远,张 飒,等.融合时间与兴趣相似度的产品推荐方法研究[J].计算机技术与发展,2019,29(09):195-199.[doi:10. 3969 / j. issn. 1673-629X. 2019. 09. 037]
　KONG Yuan-yuan,BAI Zhi-yuan,ZHANG Sa,et al.Research on Products Recommendation Method Integrated with Time Weight and Interest Similarity[J].,2019,29(09):195-199.[doi:10. 3969 / j. issn. 1673-629X. 2019. 09. 037]
点击复制

融合时间与兴趣相似度的产品推荐方法研究()
分享到：

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

卷:
29
期数:
2019年09期
页码:
195-199
栏目:
应用开发研究
出版日期:
2019-09-10

文章信息/Info

Title:
Research on Products Recommendation Method Integrated with Time Weight and Interest Similarity
文章编号:
1673-629X(2019)09-0195-05
作者:
孔元元; 白智远; 张 飒; 吕 品
上海电机学院 电子信息学院,上海 201306
Author(s):
KONG Yuan-yuan; BAI Zhi-yuan; ZHANG Sa; LYU Pin
School of Electronic Information,Shanghai Dianji University,Shanghai 201306,China
关键词:
电视产品; 协同过滤算法; 时间权重; 兴趣相似度; 个性化推荐
Keywords:
television products; collaborative filtering algorithm; time weight; interest similarity; personalized recommendation
分类号:
TP311.1
DOI:
10. 3969 / j. issn. 1673-629X. 2019. 09. 037
摘要:
互联网信息技术的发展使得企业可以为众多在线用户实现信息的实时交互。 如何挖掘出海量产品数据中隐藏的用户行为、实现个性化推荐服务是企业面临的一个重要问题。 本项目使用 PHP、Python、MariaDB 等技术对原始数据进行了清理、集成、标记等预处理,然后以用户消费信息中的产品信息为研究对象,运用传统的协同过滤算法,建立用户与产品信息的 0-1 矩阵,得到产品的偏好推荐。 通过测试推荐结果发现,模型效果欠佳。 为了提高推荐精度,提出了时间权重与兴趣相似度融合的协同过滤模型。 时间权重考虑了用户偏好变化与时间的依赖关系,兴趣相似度用于改进模型的预测精度。 在包含 4 万余条电视产品收视数据的数据集上实现了该方法,并将其与传统协同过滤模型进行了对比,发现改进后的协同过滤模型的精度得到了显著改善。 最后,基于时间权重与兴趣相似度融合的协同过滤模型的推荐结果,给出了增加用户所使用的机顶盒套餐信息的个性化营销推荐方案。
Abstract:
The development of Internet information technology has enabled enterprises to realize real-time information interaction for many online users. How to dig out hidden user behaviors in the massive products data and implement personalized recommendation services is an important issue for operators. Firstly,we use MariaDB,PHP and Python to clean,integrate,label the raw datasets. Secondly,taking the products information in the user’s consumption information as the research object,the 0-1 matrix of user and TV product information is established based on the traditional collaborative filtering algorithm,and the products preference recommendation is obtained. We found that the accuracy of the traditional collaborative filtering algorithm is not desirable by observing the test recommendation results. To improve the recommendation accuracy,an advanced collaborative filtering model integrated with time weight and interest similarity is proposed. The time weight takes into account the dependence of the user’s viewing preference changing with time,and the similarity of interest is used to improve the prediction accuracy of the proposed model. The proposed model is implemented on a data set containing more than 40 000 TV products. Finally,we compared the proposed model against the traditional collaborative filtering model. The experiment suggests that the accuracy of the proposed collaborative filtering model is significantly improved. In addition,based on recommendation results of ntegrating time weights and interest similarity,the personalized marketing recommendation scheme,with the package information of the set-top boxes added,is put forward to the decision-makers.

相似文献/References:

[1]叶树鑫[],何聚厚[][]. 协作学习中基于协同过滤的学习资源推荐研究[J].计算机技术与发展,2014,24(10):63.
　YE Shu-xin[],HE Ju-hou[][]. esearch on Learning Material Recommendation Based on Collaborative Filtering Algorithm in Cooperative Learning[J].,2014,24(09):63.
[2]谢人强,陈震. 基于共同评分项和权重计算的推荐算法研究[J].计算机技术与发展,2016,26(09):69.
　XIE Ren-qiang,CHEN Zhen. Research on Recommendation Algorithm Based on Co-rating and Weight Calculation[J].,2016,26(09):69.
[3]马婉贞,钱育蓉. 基于标签匹配的协同过滤推荐算法研究[J].计算机技术与发展,2017,27(07):25.
　MA Wan-zhen,QIAN Yu-rong. Investigation on Collaborative Filtering Recommendation Algorithm with Tag Matching[J].,2017,27(09):25.
[4]杨大鑫,王荣波,黄孝喜,等.基于最小方差的 K-means 用户聚类推荐算法[J].计算机技术与发展,2018,28(01):104.[doi:10.3969/ j. issn.1673-629X.2018.01.022]
　YANG Da-xin,WANG Rong-bo,HUANG Xiao-xi,et al.K-means User Clustering Recommendation Algorithm Based on Minimum Variance[J].,2018,28(09):104.[doi:10.3969/ j. issn.1673-629X.2018.01.022]
[5]沈鹏,李涛.混合协同过滤算法在推荐系统中的应用[J].计算机技术与发展,2019,29(03):69.[doi:10.3969/ j. issn.1673-629X.2019.03.014]
　SHEN Peng,LI Tao.Application of Hybrid Collaborative Filtering Algorithm in Recommendation System[J].,2019,29(09):69.[doi:10.3969/ j. issn.1673-629X.2019.03.014]
[6]马瑞新,郭芳清,刘振娇,等.融合上下文信息与核密度估计的协同过滤推荐[J].计算机技术与发展,2021,31(04):34.[doi:10. 3969 / j. issn. 1673-629X. 2021. 04. 006]
　MA Rui-xin,GUO Fang-qing,LIU Zhen-jiao,et al.Collaborative Filtering Recommendation Algorithm for FusionContext Information and Kernel Density Estimation[J].,2021,31(09):34.[doi:10. 3969 / j. issn. 1673-629X. 2021. 04. 006]
[7]任静霞,武志峰.动态信任衰减和信息匹配的混合推荐算法[J].计算机技术与发展,2021,31(10):30.[doi:10. 3969 / j. issn. 1673-629X. 2021. 10. 006]
　REN Jing-xia,WU Zhi-feng.Hybrid Recommendation Algorithm Based on Dynamic Trust Decay and Information Matching[J].,2021,31(09):30.[doi:10. 3969 / j. issn. 1673-629X. 2021. 10. 006]
[8]刘昊东,王 诚.基于热门度修正因子和置信度的协同过滤算法[J].计算机技术与发展,2023,33(03):127.[doi:10. 3969 / j. issn. 1673-629X. 2023. 03. 019]
　LIU Hao-dong,WANG Cheng.Collaborative Filtering Algorithm Based on Popularity Correction Factor and Confidence[J].,2023,33(09):127.[doi:10. 3969 / j. issn. 1673-629X. 2023. 03. 019]

更新日期/Last Update: 2019-09-10