基于网页概率潜在语义信息的用户兴趣聚类

J4 ›› 2014, Vol. 36 ›› Issue (04): 765-771.

基于网页概率潜在语义信息的用户兴趣聚类

钱雪忠，吴志媛

(江南大学物联网工程学院，江苏无锡 214122)

收稿日期:2012-09-24 修回日期:2013-03-29 出版日期:2014-04-25 发布日期:2014-04-25
基金资助:
国家自然科学基金资助项目(61103129)；江苏省科技支撑计划资助项目(BE2009009)

User’s interest clustering based on
webpage probabilistic latent semantic information

QIAN Xuezhong,WU Zhiyuan

(School of Internet of Things Engineering,Jiangnan University,Wuxi 214122,China)

Received:2012-09-24 Revised:2013-03-29 Online:2014-04-25 Published:2014-04-25

摘要/Abstract

摘要：

为了能准确挖掘用户兴趣点，首先利用概率潜在语义分析PLSA模型将“网页词”矩阵向量投影到概率潜在语义向量空间，并提出“自动相似度阈值选择”方法得到网页间的相似度阈值，最后提出将平面划分法与凝聚式层次聚类相结合的凝聚式层次k中心点HAKmedoids算法，实现用户兴趣点聚类。实验结果表明,与传统的基于划分的算法相比，HAKmedoids算法聚类效果更好。同时，提出的用户兴趣点聚类技术在个性化服务领域可提高个性化推荐和搜索的效率。关键词：

关键词: 概率潜在语义分析, 自动相似度阈值选择, 用户兴趣点, 凝聚式层次k中心点, 个性化服务

Abstract:

To mine user’s interests accurately,probabilistic latent semantic analysis (PLSA) model is firstly used to project webpage-word matrix vector into probabilistic latent semantic vector space. A method of “auto-selected similarity threshold” is proposed to get web pages similarity threshold. At last, combined with divisiory algorithms and hierarchical agglomerative clustering,a hierarchical agglomerative kmedoids clustering algorithm is proposed to realize cluster user’s interests. The experimental results show that, compared with the traditional divisiory algorithms, the hierarchical agglomerative kmedoids algorithm has a better clustering effect. Furthermore, user’s interest clustering technique can improve the efficiency of personalized recommendation and search in user’ personalized service fields.

Key words: probabilistic latent semantic analysis;autoselected similarity threshold;user’s interest points;hierarchical agglomerative kmedoids;personalized service

钱雪忠，吴志媛. 基于网页概率潜在语义信息的用户兴趣聚类[J]. J4, 2014, 36(04): 765-771.

QIAN Xuezhong,WU Zhiyuan. User’s interest clustering based on
webpage probabilistic latent semantic information [J]. J4, 2014, 36(04): 765-771.

参考文献

［1］Xiao Sheng, Hu Jinzhu, Yao Shuangyun, et al. Study on feature item extraction method based on ontology view ［J］. Application Research of Computers,2010, 27(1):4244.(in Chinese)
［2］Zeng Chun,Xing Chunxiao,Zhou Lizhu. A survey of personalization technology ［J］. Journal of Software,2002,13(10):19521961.(in Chinese)
［3］Zhu Zhengyu,Zhang Xiaolin,Xiong Qian,et al. An algorithm of collaborative recommendation based on user’s interest subclass ［J］.Computer Science,2005, 32(10):176180.(in Chinese)
［4］Pan Yanjun. Personalized research combining with analysis of web user’s behavior based on the user’s behavior browsing content ［D］. Tianjin:Tianjin University,2005.(in Chinese)
［5］Hsieh Yachao, Huang Yutsun, Wang ChienChih, et al. Improved spoken document retrieval with dynamic key term lexicon and probabilistic latent semantic analysis(PLSA)［C］∥Proc of Acoustics, Speech and Signal Processing, 2006:961964.
［6］Li Sheng,Hu Heping. An effective retrieval method based on probabilistic latent semantic analysis ［J］. Journal of Huazhong University of Science & Technology(Natural Science Edition),2010,38(11):4850.(in Chinese)
［7］Li Yuanyuan, Ma Yongqiang. Text term weighting approach based on latent semantic indexing ［J］. Computer Applications,2008,28(6):14601462.(in Chinese)
［8］Bhat V, Oates T, Shanbhag V, et al. Finding aliases on the web using latent semantic analysis ［J］.Data &Knowledge Engineering , 2004, 49(2):129143.
［9］Chen Shuran.Personlizationbased user interest modeling and its applying study［D］. Chongqing:Chongqing University,2007.(in Chinese)
［10］Zhang Meng,Wang Daling,Yu Ge. A text clustering method based on autoselected threshold［J］. Journal of Computer Research and Development,2004,10(41):17481753.(in Chinese)
［11］Ma Suqin,Shi Huaji.Text density clustering algorithm with optimized threshold values ［J］. Computer Engineering and Applications,2011,47(17):134136.(in Chinese)
［12］Jia Ruiyu,Geng Jinwei,Ning Zaizao,et al. Fast clustering algorithm based on representative points ［J］. Computer Engineering and Applications,2010, 46(33):121123.(in Chinese)
［13］Guo Jingfeng,Zhao Yuyan,Bian Weifeng,et al. A hierarchical clustering algorithm based on improved cluster cohesion and separation［J］. Journal of Computer Research and Development,2008,45(Z1):202206. (in Chinese)
［14］Zhang Yufang,Zhu Jun,Xiong Zhongyang. Improved text clustering algorithm of probabilistic latent with semantic analysis ［J］. Journal of Computer Applications,2011, 31(3):674676.(in Chinese)
［15］Hochul J,Taehwan K,Joongmin C.Adaptive user profiling for personalized information retrieval［C］∥Proc of the 3rd International Conference on Convergence and Hybrid Information Technology, 2008:836841.
附中文参考文献:
［1］肖升,胡金柱,姚双云,等.基于本体视图特征项抽取方法研究［J］.计算机应用研究,2010, 27(1):4244.
［2］曾春,邢春晓,周立柱.个性化服务技术综述［J］.软件学报,2002,13(10):19521961.
［3］朱征宇,张小林,熊茜,等.基于用户兴趣子类的协作推荐算法［J］.计算机科学,2005, 32(10):176180.
［4］潘延军.基于用户浏览内容的Web用户浏览行为个性化研究［D］.天津：天津大学,2005.
［6］李胜,胡和平.一种基于PLSA的高效检索方法［J］.华中科技大学学报(自然科学版),2010,38(11):4850.
［7］李媛媛,马永强.基于潜在语义索引的文本特征词权重计算方法［J］.计算机应用,2008,28(6):14601462.
［9］陈抒然.面向个性化服务的用户兴趣建模及应用研究［D］.重庆：重庆大学,2007.
［10］张猛,王大玲,于戈.一种基于自动阈值发现的文本聚类方法［J］.计算机研究与发展,2004,10(41):17481753.
［11］马素琴,施化吉.阈值优化的文本密度聚类算法［J］.计算机工程与应用,2011,47(17):134136.
［12］贾瑞玉,耿锦威,宁再早,等.基于代表点的快速聚类算法［J］.计算机工程与应用,2010, 46(33):121123.
［13］张玉芳,朱俊,熊忠阳.改进的概率潜在语义分析下的文本聚类算法［J］.计算机应用,2011, 31(3):674676.
［14］郭景峰,赵玉艳,边伟峰,等.基于改进的凝聚性和分离性的层次聚类算法［J］.计算机研究与发展,2008,45(Z1):202206.

[1]	肖来元，殷明. 基于身份和上下文的个性化服务研究[J]. J4, 2012, 34(11): 28-33.
[2]	刘毅. 一种图像局部特征的语义提取方法[J]. J4, 2010, 32(6): 61-64.
[3]	俞辉. 基于PLSA模型的Web用户聚类算法研究[J]. J4, 2008, 30(7): 30-32.
[4]	刘星沙[1] 谭利球[1] 熊拥军[2]. 关联规则挖掘算法及其应用研究[J]. J4, 2007, 29(1): 83-85.
[5]	杨清[1,2] 游星雅[1] 蒋向红[1]. 基于LSI的图书馆个性化信息服务系统的设计与研究[J]. J4, 2006, 28(4): 15-18.
[6]	付关友朱征宇. 个性化服务中基于行为分析的用户兴趣建模[J]. J4, 2005, 27(12): 76-78.
[7]	应晓敏刘明窦文华. 一种面向个性化服务的客户端细粒度用户建模方法[J]. J4, 2003, 25(6): 39-42.