| 恐怖事件情境下微博信息与评论用户的画像及比较 |
| |
| 引用本文： | 安璐,周亦文.恐怖事件情境下微博信息与评论用户的画像及比较[J].情报科学,2020,38(4):9-16. |
| |
| 作者姓名： | 安璐 周亦文 |
| |
| 作者单位： | 武汉大学信息资源研究中心,湖北武汉430072;武汉大学信息管理学院,湖北武汉430072;武汉大学信息管理学院,湖北武汉430072 |
| |
| 基金项目： | 教育部哲学社会科学研究重大课题攻关项目“提高反恐怖主义情报信息工作能力对策研究”(17JZD034);国家自然科学基金青年项目“突发公共卫生事件社交媒体信息主题演化与影响力建模”(71603189);国家自然科学基金重大课题“国家安全大数据综合信息集成与分析方法”(71790612)。 |
| |
| 摘 要： | 【目的/意义】构建用户特征指标体系,对恐怖事件情境下微博信息与评论用户进行画像并比较有助于掌握参与恐怖事件讨论用户的特点,加强反恐舆论引导。【方法/过程】以"#巴塞罗那恐怖袭击#"话题下的所有微博及评论数据为例,使用基于相关性的LDA主题模型提取微博主题,从用户特征和文本特征两个角度构建指标体系,并采用两步聚类刻画微博用户特征,分析发布微博用户和评论用户的异同。【结果/结论】以往活跃度、影响力较高的用户在该事件中不一定拥有较高的影响力;原始微博用户的平均等级略低于评论用户,但其在该事件中的影响力高于评论用户;原始微博用户类型多于评论微博用户类型。
|
| 关 键 词： | 用户画像 微博 评论 生命周期 主题分析 聚类 恐怖事件 |
The Profiling and Comparison between Microblogging Information and Comment Users in the Context of Terrorist Events |
| |
| AN Lu,ZHOU Yi-wen.The Profiling and Comparison between Microblogging Information and Comment Users in the Context of Terrorist Events[J].Information Science,2020,38(4):9-16. |
| |
| Authors: | AN Lu ZHOU Yi-wen |
| |
| Affiliation: | (Center for Studies of Information Resources,Wuhan University,Wuhan 430072,China;School of Information Management,Wuhan University,Wuhan 430072,China) |
| |
| Abstract: | 【Purpose/significance】The study aims to construct the feature index system of users and to profile the users who post and those who comment on the microblogging platform in the context of terrorism events. It can illustrate the characteristics of users who participated in the discussion of terrorist attacks and would be helpful to guide the public opinion of counter-terrorism.【Method/process】This study investigated the microblogging entries and their comments related to the topic of # Barcelona terrorist attack# and extracted the topics by relevance-based Latent Dirichlet Allocation(LDA) model.The index system was constructed from the perspectives of user features and text features. The two-step clustering technique was used to characterize the users. The similarities and differences between the users who posted and those who commented were analyzed.【Result/conclusion】It was found that users with relatively high activity and influence in the past may not necessarily have high influence in this event. The average rating of microblogging users who posted was slightly lower than that of commenting users. However, the former were more active and had a higher influence than the latter in the event. The types of users who posted were more diverse than those who commented. |
| |
| Keywords: | user profile microblog comment life cycle topical analysis clustering terrorist event |
| 本文献已被 维普 万方数据 等数据库收录！ |
| 点击此处可从《情报科学》浏览原始摘要信息 |
|
点击此处可从《情报科学》下载全文 |
