微博自媒体账号识别研究

发布时间：2018-05-18 20:46

本文选题：微博 + 自媒体账号　；参考：《计算机研究与发展》2015年11期

【摘要】：随着Web 2.0时代的发展,微博作为新兴的社交网络媒体在人们的日常生活中扮演着愈发重要的角色.它不仅是用户交流与分享信息的桥梁,也是获取信息的重要方式.微博同时具有社交网络与信息媒体双重性,其生态环境中仅具有媒体属性,用于发布信息给公众的自媒体账号(we media account)发展迅速.首次提出微博自媒体账号识别这一研究问题,阐述了自媒体账号识别对分析微博生态环境、用户兴趣建模、优质内容挖掘的重要意义,提出了结合个人信息、账号行为及微博内容3类特征的有监督识别方法.研究结果表明:1)自媒体账号与普通的微博账号有着较明显的不同,主要体现在微博发布行为的规律性以及话题分布特性之上.2)提出的3类特征能够有效识别自媒体账号,不同类别的特征也能够相互补充,预测准确率高达96.71%.
[Abstract]:With the development of Web 2.0 era, Weibo, as a new social network media, plays an increasingly important role in people's daily life. It is not only a bridge for users to exchange and share information, but also an important way to obtain information. Weibo has the dual nature of social network and information media, and its ecological environment has only media attributes, and the self-media account for publishing information to the public has developed rapidly. This paper first puts forward the research problem of Weibo self-media account recognition, expounds the importance of self-media account identification in analyzing Weibo ecological environment, user interest modeling and high-quality content mining, and puts forward a combination of personal information. The supervised identification method of account behavior and Weibo content. The results show that: 1) the self-media account is obviously different from the ordinary Weibo account, which is mainly reflected in the regularity of Weibo publishing behavior and the characteristics of topic distribution. 2) the three kinds of features proposed by the author can effectively identify the self-media account. Different types of features can also complement each other, forecasting accuracy as high as 96.71.
【作者单位】：北京大学信息科学技术学院网络与信息系统研究所;
【基金】：国家自然科学基金项目(61272343) 教育部高等学校博士学科点专项科研基金项目(20130001110032)
【分类号】：TP393.092

【参考文献】