|
成为一个效应而不仅是隐喻：从普及到社会联动
To Be a Star Is Not Only Metaphoric: From Popularity to Social Linkage |
|
| |
| 课程网址: |
http://videolectures.net/icwsm2010_couronne_tbs/
|
| 主讲教师: |
Thomas Couronne
|
| 开课单位: |
法国电信研究
|
| 开课时间: |
信息不详。欢迎您在右侧留言补充。
|
| 课程语种: |
英语
|
| 中文简介: |
在线平台的出现允许将自我出版活动和社交网络相结合，为建立在线声誉和可见性提供了新的可能性。本文提出了一种分析网络人气的方法，该方法既考虑了发布内容的成功，又考虑了社交网络拓扑结构。首先，我们对Kohonen自组织地图进行了调整，以便根据用户和权威特征对在线平台的用户进行集群。然后，我们对节点在社交网络中的组织方式进行了详细的分析。最后，我们研究了每个节点周围的网络局部结构与相应用户的人气之间的关系。我们将此方法应用于MySpace音乐社交网络。我们观察到，最受欢迎的艺术家是星型社会结构的中心，它存在于参与社区和社会活动动态的艺术家中的一小部分，而这些艺术家的受欢迎程度与他们的受欢迎程度无关。这种基于学习算法和网络分析的方法对于丰富的在线行为描述似乎是一种强大而直观的技术。
|
| 课程简介: |
The emergence of online platforms allowing to mix self publishing activities and social networking offers new possibilities for building online reputation and visibility. In this paper we present a method to analyze the online popularity that takes into consideration both the success of the published content and the social network topology. First, we adapt the Kohonen self organizing maps in order to cluster the users of online platforms depending on their audience and authority characteristics. Then, we perform a detailed analysis of the manner nodes are organized in the social network. Finally, we study the relationship between the network local structure around each node and the corresponding user’s popularity. We apply this method to the MySpace music social network. We observe that the most popular artists are centers of star shaped social structures and that it exists a fraction of artists who are involved in community and social activity dynamics independently of their popularity. This method based on a learning algorithm and on network analysis appears to be a robust and intuitive technique for a rich description of the online behavior.
|
| 关 键 词: |
社交网络; 学习算法; 网络分析
|
| 课程来源: |
视频讲座网
|
| 最后编审: |
2019-11-22：cwx
|
| 阅读次数: |
31
|
|
图 书 馆