基于HDP模型的领域微博主题演化研究

计算机工程

基于HDP模型的领域微博主题演化研究

高永兵¹,杨利莹¹,胡文江¹,马占飞²

(1.内蒙古科技大学信息工程学院,内蒙古包头 014010; 2.包头师范学院信息工程系,内蒙古包头 014010)

收稿日期:2017-02-23 出版日期:2018-02-15 发布日期:2018-02-15
作者简介:高永兵(1974—),男,副教授、硕士,主研方向为计算语言学、Web数据管理;杨利莹,硕士研究生;胡文江,教授;马占飞,教授、博士。
基金资助:
国家自然科学基金(61163025);内蒙古自治区自然科学基金(2015MS0621)。

Research on Domanial Microblog Topic Evolution Based on HDP Model

GAO Yongbing ¹,YANG Liying ¹,HU Wenjiang ¹,MA Zhanfei ²

(1.College of Information Engineering,Inner Mongolia University of Science and Technology,Baotou,Inner Mongolia 014010,China;2.Department of Information Engineering,Baotou Teachers’ College,Baotou,Inner Mongolia 014010,China)

Received:2017-02-23 Online:2018-02-15 Published:2018-02-15

摘要/Abstract

摘要： 领域微博中包含较多的专业领域信息,并且随时间表现出较强的演化性。为分析领域的主题演化情况,构建一个基于分层Dirichlet过程(HDP)的DM-HDP模型。以用户为单位抽取领域相关的微博,利用微博的领域特征和时间特征,提取领域相关带有明显时间特征的微博并自动挖掘其主题分布,最终构建领域主题演化分析过程。实验结果表明,基于DM-HDP模型的分析方法能够表现领域微博主题的演化过程,与基于LDA和HDP模型的方法相比,在内容困惑度和模型复杂度等方面均具有明显优势。

关键词: 领域微博, 主题挖掘, 分层Dirichlet模型, DM-HDP模型, Gibbs采样, 主题演化

Abstract: Domanial microblog contains much professional information that show a strong evolution over time.In order to analyze the topics of professional microblog automatically,a domanial microblog topic evolution method based on Hierarchical Dirichlet Process(HDP) model is built.Firstly,domain-related microblog is extracted with the individual user as the unit.Then,accurate extraction of domain-related microblog with distinct temporal features and automatic mining of its topics using domain features and temporal features.At last,the process of domanial topics evolution analysis is constructed.Experimental results show that the method based on the DM-HDP model can show the evolution of the field of microblog,and compared with the methods that based on the LDA and HDP model,it has obvious advantages in terms of content confusion and model complexity.

Key words: domanial microblog, topic mining, hierarchical Dirichlet model, DM-HDP model, Gibbs sampling, topic evolution

中图分类号:

TP391

高永兵,杨利莹,胡文江,马占飞. 基于HDP模型的领域微博主题演化研究[J]. 计算机工程.

GAO Yongbing,YANG Liying,HU Wenjiang,MA Zhanfei. Research on Domanial Microblog Topic Evolution Based on HDP Model[J]. Computer Engineering.

/ / 推荐 / 导出引用

链接本文: http://www.ecice06.com/CN/Y2018/V44/I2/1

参考文献

参考文献［1］JENSEN S,LIU X,YU Y,et al.Generation of Topic Evolution Trees from Heterogeneous Bibliographic Networks［J］.Journal of Informetrics,2016,10(2):606-621. ［2］叶春蕾,冷伏海.基于社会网络分析的技术主题演化方法研究［J］.情报理论与实践,2014,37(1):126-130. ［3］陈千,桂志国,郭鑫,等.基于特征本体的文本流主题演化［J］.计算机应用,2015,35(2):456-460. ［4］MA J,SUN M,LI C,et al.Ontology Evolution Algorithm for Topic Information Collection［J］.International Journal of Nonlinear Science,2014,18(1):86-91. ［5］BLEI D M,NG A Y,JORDAN M I.Latent Dirichlet Allocation［J］.Journal of Machine Learning Research,2003,32(3):993-1022. ［6］WANG Y,AGICHTEIN E,BENZI M.TM-LDA:Efficient Online Modeling of Latent Topic Transitions in Social Media［C］//Proceedings of ACM SIGKDD Inter-national Conference on Knowledge Discovery and Data Mining.New York,USA:ACM Press,2012:123-131. ［7］胡吉明,陈果.基于动态LDA主题模型的内容主题挖掘与演化［J］.图书情报工作,2014,58(2):138-142. ［8］HOFFMAN M D,BLEI D M,BACH F R.Online Learning for Latent Dirichlet Allocation［J］.Advances in Neural Information Processing Systems,2010,23(5):856-864. ［9］TEH Y W,JORDAN M I,BEAL M J,et al.Hierarchical Dirichlet Processes［J］.Journal of the American Statistical Association,2006,101(476):1566-1581. ［10］BLEI D M,LAFFERTY J D.Dynamic Topic Models［C］// Proceedings of the 23rd International Conference on Machine Learning.Washington D.C.,USA:IEEE Press,2006:113-120. ［11］AHMED A,XING E P.Dynamic Non-parametric Mixture Models and the Recurrent Chinese Restaurant Process:with Applications to Evolutionary Clustering［C］//Proceedings of Siam International Conference on Data Mining.Atlanta,USA:SDM Press,2008:219-230. ［12］ZHANG J,SONG Y,ZHANG C,et al.Evolutionary Hierarchical Dirichlet Processes for Multiple Correlated Time-varying Corpora［C］//Proceedings of ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.New York,USA:ACM Press,2010:1079-1088. ［13］HONG L,DOM B,GURUMURTHY S,et al.A Time-dependent Topic Model for Multiple Text Streams［C］//Proceedings of ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.New York,USA:ACM Press,2011:832-840. ［14］方莹,黄河燕,辛欣,等.面向动态主题数的话题演化分析［J］.中文信息学报,2014,28(3):142-149. ［15］FU X,LI J,YANG K,et al.Dynamic Online HDP Model for Discovering Evolutionary Topics from Chinese Social Texts［J］.Neurocomputing,2015,171(C):412-424. ［16］张晨逸,孙建伶,丁轶群.基于MB-LDA模型的微博主题挖掘［J］.计算机研究与发展,2011,48(10):1795-1802. ［17］刘少鹏,印鉴,欧阳佳,等.基于MB-HDP模型的微博主题挖掘［J］.计算机学报,2015,38(7):1408-1419. ［18］FERGUSON T S.A Bayesian Analysis of Some Nonparametric Problems［J］.Annals of Statistics,1973,1(2):209-230. ［19］周建英,王飞跃,曾大军.分层Dirichlet过程及其应用综述［J］.自动化学报,2011,37(4):389-407. ［20］陈晓伟.基于主题爬虫与文本分类的微博资讯智能生成策略研究［D］.武汉:华中科技大学,2013. ［21］张力生,年欢,宋辉,等.领域模型中关联语义的描述逻辑表示与应用［J］.软件,2015(6):66-74. ［22］MENG X,WEI F,LIU X,et al.Entity-centric Topic-oriented Opinion Summarization in Twitter［C］//Proceedings of ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.New York,USA:ACM Press,2012:379-387. 编辑吴云芳

[1]	许伟佳, 秦永彬, 黄瑞章, 陈艳平. 基于DMA与特征划分的多源文本主题模型[J]. 计算机工程, 2021, 47(7): 59-66.
[2]	姚立,张曦煌. 一种基于标签的改进主题演化模型[J]. 计算机工程, 2019, 45(4): 205-210,216.
[3]	李又玲,常致全. 基于Gibbs采样与概率分布估计的移动云数据存储[J]. 计算机工程, 2017, 43(1): 13-19.
[4]	李博,陈志刚,黄瑞,郑祥云. 基于LDA模型的音乐推荐算法[J]. 计算机工程, 2016, 42(6): 175-179,184.
[5]	刘良选,黄梦醒. 一种面向词汇突发的连续时间主题模型[J]. 计算机工程, 2016, 42(11): 195-201.
[6]	李敬,印鉴,刘少鹏,潘雅丽. 基于话题标签的微博主题挖掘[J]. 计算机工程, 2015, 41(4): 30-35.
[7]	刘文远, 田陆芳, 王常武, 王宝文. 基于Gibbs采样与遗传算法的模体识别[J]. 计算机工程, 2011, 37(14): 180-182.

选择文件类型/文献管理软件名称

选择包含的内容