Microblog Hot Topic Evolution Based on Improved On-Line Biterm Topic Model

doi:10.3778/j.issn.1002-8331.2007-0151

Abstract

Abstract:

Topic evolution analysis is one of the research hotspots of public opinion monitoring. The evolution analysis of microblog hot topics is of great practical significance to network users and network regulators. To solve the problem of OBTM topic mixing and high probability of redundant words, the OBTM based on topic labels and prior parameters （LPOBTM） is proposed in this paper. According to the topic labels, the microblog text set is divided into two types of data sets with and without topic labels. Different document-topic prior parameters are set. Based on document-topic probability distribution in the previous time slice, the intensity ranking of all topics is carried out by drawing lessons from the Sigmod function. Thus, the prior parameter calculation method of topic-word distribution on current time slice is optimized. The experimental results show that LPOBTM can describe the content evolution of topics more accurately, and has lower model perplexity.

Key words: topic label, prior parameter, topic intensity ranking, On-line Biterm Topic Model（OBTM）, microblog hot topic evolution

摘要：

话题演化分析是舆情监控的研究热点之一，面向微博热点话题进行演化分析，对于网络用户以及网络监管部门都有很重要的现实意义。针对在线词对主题模型（On-line Biterm Topic Model，OBTM）新旧主题混合、冗余词概率相对较高的问题，对OBTM进行改进，提出基于话题标签和先验参数的OBTM模型（Topic Labels and Prior Parameters OBTM，LPOBTM）。根据微博热点话题的话题标签，将微博文本集区分为含话题标签和不含话题标签的两类数据集，并设置不同的文档-主题先验参数；在前一时间片文档-主题概率分布的基础上，借鉴Sigmod函数对所有主题进行强度排名，从而优化当前时间片上主题-词分布的先验参数计算方法。实验结果表明，LPOBTM能够更准确地描述话题的内容演化情况，并且有更低的模型困惑度。

关键词: 话题标签, 先验参数, 主题强度排名, 在线词对主题模型, 微博热点话题演化

WU Di, ZHANG Mengtian, SHENG Long, HUANG Zhuyun, GU Mingxing. Microblog Hot Topic Evolution Based on Improved On-Line Biterm Topic Model[J]. Computer Engineering and Applications, 2021, 57(24): 179-184.

吴迪，张梦甜，生龙，黄竹韵，顾明星. 改进在线词对主题模型的微博热点话题演化[J]. 计算机工程与应用, 2021, 57(24): 179-184.

References

[1] 李鹏飞，董旭，仲兆满，等.基于微博用户兴趣话题的相似用户挖掘[J].计算机工程与应用，2019，55（11）：102-109.
LI P F，DONG X，ZHONG Z M，et al.Similar user mining based on user interest topics in Weibo[J].Computer Engineering and Applications，2019，55（11）：102-109.
[2] 吴晓娟.基于微博文本的网络舆情主题演化分析[D].南京：南京大学，2018.
WU X J.Analysis of the theme evolution of internet public opinion based on microblog text[D].Nanjing：Nanjing University，2018.
[3] 陈红阳，汪林林，鲁江坤，等.基于双态模型的微博话题跟踪方法研究[J]，计算机工程与应用，2017，53（16）：144-148.
CHEN H Y，WANG L L，LU J K，et al.Research on method of topic tracking for micro-blog texts based on double topic model[J].Computer Engineering and Applications，2017，53（16）：144-148.
[4] WANG X，MCCALLUM A.Topics over time：a non-Markov continuous-time model of topical trends[C]//ACM SIGKDD International Conference on Knowledge Discovery and Data Mining，2006：424-433.
[5] BLEI D M，LAFFERTY J D.Dynamic topic models[C]//Proceedings of International Conference on Machine Learning，2006：113-120.
[6] 王子涵，王玉辉，王雷，等.一种基于社交媒体的突发事件话题演化分析系统研究[J].信息网络安全，2017（9）：98-102.
WANG Z H，WANG Y H，WANG L，et al.Research on a topic evolution analysis system of emergency based on social media[J].Netinfo Security，2017（9）：98-102.
[7] ALSUMAIT L，DOMENICONI C.Online LDA：adaptive topic models for mining text streams with applications to topic detection and tracking[C]//Eighth IEEE International Conference on Data Mining，2008：3-12.
[8] YAN X，GUO J，LAN Y，et al.A biterm topic model for short texts[C]//Proceedings of the 22nd International Conference on World Wide Web，2013：1445-1456.
[9] 裴可锋，陈永洲，马静.基于OLDA的可变在线主题演化模型[J].情报科学，2017（5）：66-71.
PEI K F，CHEN Y Z，MA J.Variable online theme evolution model based on OLDA[J].Information Science，2017（5）：66-71.
[10] 蒋权，郑山红，刘凯，等.DOLDA模型设计与主题演化分析[J].计算机工程与设计，2018，39（2）：446-451.
JIANG Q，ZHENG S H，LIU K，et al.Design of DOLDA model and analysis of theme evolution[J].Computer Engineering and Design，2018，39（2）：446-451.
[11] 余本功，张卫春，王龙飞.基于改进的OLDA模型话题检测及演化分析[J].情报杂志，2017，36（2）：102-107.
YU B G，ZHANG W C，WANG L F.Topic detection and evolution analysis based on improved OLDA model[J].Journal of Intellgence，2017，36（2）：102-107.
[12] 李慧，王丽婷.基于话题标签的微博热点话题演化研究[J].情报科学，2019，37（1）：30-36.
LI H，WANG L T.Microblog hot topic evolution research based on topic label[J].Information Science，2019，37（1）：30-36.
[13] 苏英杰.微博话题演化追踪算法的设计与实现[D].昆明：云南大学，2018.
SU Y J.The design and implementation of topic evolution tracking for microblog[D].Kunming：Yunnan University，2018.
[14] 徐健，吴思洋.网络用户评论的情感分歧度量化算法研究[J].情报学报，2020，39（4）：427-435.
XU J，WU S Y.The quantitative algorithm of sentiment divergence based on Web user reviews[J].Journal of the China Society for Scientific and Technical Information，2020，39（4）：427-435.
[15] 晏小辉.短文本话题建模[D].北京：中国科学院大学，2014.
YAN X H.Topic modeling over short texts[D].Beijing：University of Chinese Academy of Sciences，2014.
[16] ZHAO W，CHEN J，PERKINS R，et al.A heuristic approach to determine an appropriate number of topics in topic modeling[J].BMC Bioinformatics.BioMed Central，2015，16（13）：S8.