ISSN: 1673-825X    Imprint: Chongqing University of Posts and Telecommunications Journal
Volume listings
  • Current volume
  •  
  • Search by keywords:
  • Search by 
  • Keywords
  • To
  • Journal archive
Related Links
基于微博多维度及综合权值的热点话题检测模型
Hot topic detection model based on multi-dimensions and comprehensive weights of microblog
DOI:10.3979/j.issn.1673-825X.2019.04.006
Received:November 13, 2018  Revised:April 15, 2019
中文关键词:网络舆情  微博  热点话题  综合权值  走势分析
英文关键词:internet public opinion  microblog  hot topics  comprehensive weight  trend analysis
基金项目:国家社会科学基金西部项目(17XFX013); 重庆市科学技术委员会技术创新与应用示范项目(cstc2018jscx-msybX0332)
Author NameAffiliationE-mail
CHENG Kefei Key Laboratory of Network Intelligence and Network Technology, Chongqing University of Posts and Telecommunications, Chongqing 400065, P. R. China chengkf@cqupt.edu.cn 
DENG Xianjun Key Laboratory of Network Intelligence and Network Technology, Chongqing University of Posts and Telecommunications, Chongqing 400065, P. R. China 406455850@qq.com 
ZHOU Ke Network Security and Defense Corps, Chongqing Municipal Public Security Bureau, Chongqing 401120, P. R. China zkheartboy@qq.com 
LUO Zhao Key Laboratory of Network Intelligence and Network Technology, Chongqing University of Posts and Telecommunications, Chongqing 400065, P. R. China 240860747@qq.com 
CHEN Xudong Key Laboratory of Network Intelligence and Network Technology, Chongqing University of Posts and Telecommunications, Chongqing 400065, P. R. China 996468223@qq.com 
Hits: 40
Download times: 20
中文摘要:
      传统热点检测算法仅从单一的某个维度衡量话题的热度,导致热点话题检测精度低,在对突发性热点话题进行检测时尤为明显。针对此问题,提出一种多维度热点话题度量模型。该模型对话题进行筛选,得到一个热点话题初始集,再融入话题热度的影响力因子,计算各个话题的综合权值,将话题的综合权值按照一定的权重与多维度热点话题度量模型进行有效融合,得到一种基于微博多维度及综合权值的热点话题检测模型。通过使用真实的微博数据进行实验对比分析,实验结果表明,提出的多维度热点话题度量模型在对突发性热点话题的检测中,其准确率(Precision)、召回率(Recall rate)和F1值(F-measure)3个评估指标相比传统算法有了较大提高;利用该模型对突发性热点话题进行跟踪,通过与官方指数进行对比,该模型能有效跟踪其发展趋势。
英文摘要:
      The traditional hotspot detection algorithm only measures the heat of a topic from a single dimension, which results in low detection accuracy of hot topics, especially when detecting sudden hot topics. To solve this problem, this paper proposes a multi-dimensional hot topic measurement model. Firstly, this model filters these topics and obtains a hot topic initial set, then integrates the influence factors of the topic heat, and calculates the comprehensive weight of each topic. And then, the comprehensive weight of the topic is effectively combined with the multi-dimensional hot topic measurement model according to a certain weight, and finally a hot topic detection model based on microblog multi-dimensional and comprehensive weight is obtained. Through the use of real microblog data for experimental comparative analysis, the experimental results show that the model proposed in this paper has a significant improvement in its accuracy, recall rate and F measure compared to the traditional algorithms in the detection of sudden hot topics. Secondly, this model is used to track sudden hot topics and it can effectively track its development trend compared with the official index.
HTML    View PDF   View/Add Comment  Download reader