Community Size Estimation of Internet Forum by Posted Article Distribution
Community Size Estimation of Internet Forum by Posted Article Distribution
复制标题
DOI：
10.1007/978-3-642-23190-2_16
复制
发表时间：
2012
期刊：
影响因子：
--
通讯作者：
M. Kubo;K. Naruse;Hiroshi Sato
中科院分区：
文献类型：
--
作者：
M. Kubo;K. Naruse;Hiroshi Sato
任何人都可以通过列举ID来轻松了解向社交媒体网站（例如博客、互联网论坛和公告板）提交消息、文章和评论的人数。然而，很难检测到一个社区的人口，其中包括对这样的网站感兴趣但没有提交任何消息的潜伏者。这不是一个简单的任务，即使是指社交媒体服务器上的日志数据。潜伏者的数量对于许多应用程序来说都是非常重要的信息。在本文中，提出了一种新的方法，基于复杂系统的动力学，估计人口的社会媒体网站，包括潜伏者。在最近的互联网论坛的研究表明，消息分布往往遵循幂律和所提出的方法利用这一事实。此方法只需要每个用户发布的消息数，因此，所有读者都可以使用此方法。此外，不需要任何特定的语言知识。在本文中，所提出的方法是由两个实际的实验证实。
Anyone can easily know the number of people who submit messages, articles, and comments to a social media web site, for example, blogs, Internet forums, and bulletin boards, by enumerating the IDs. However, it is very difficult to detect the population of a community that includes lurkers who are interested in such a web site but have not submitted any messages. It is not simple task, even if one refers to the log data on the servers for social media. The population of lurkers is very important information for many applications. In this paper, a new method, based on the dynamics of complex systems, is proposed to estimate the population of a social media site that includes lurkers. As shown in recent research of Internet forums, the message distribution tends to follow a power law and the proposed method utilizes this fact. This method only needs the number of messages posted per user, and, therefore, all readers can use this method. Moreover, one does not need any language-specific knowledge. In this paper, the proposed method is confirmed by two practical experiments.