Using Network Structure to Identify Groups in Virtual Worlds
Using Network Structure to Identify Groups in Virtual Worlds
复制标题
使用网络结构识别虚拟世界中的群体
DOI：
--
复制
发表时间：
2021
期刊：
影响因子：
--
通讯作者：
G. Sukthankar
中科院分区：
文献类型：
--
作者：
Fahad Shah;G. Sukthankar
人类是善于社交的动物，能够从语言线索和社交网络模式的组合中识别出友谊群体。但更重要的是，人们所说的内容还是他们的社会交往史？此外，是否有可能仅仅从一般的网络属性（如中心性和潜在社区的度量）来识别人们是否属于一个成员不断变化的群体？在本文中，我们解决的问题，确定社会群体的会话数据和目前的实证研究结果确定在虚拟世界中的群体。虚拟世界之所以有趣，是因为群体成员更多地由共同利益决定，而较少受到文化和社会经济因素的影响。我们的发现是，网络测量的组合比语言线索更能预测群体成员资格，并且这两种类型的特征可以结合起来提高预测能力。
Humans are adept social animals capable of identifying friendship groups from a combination of linguistic cues and social network patterns. But what is more important, the content of what people say or their history of social interactions? Moreover, is it possible to identify whether people are part of a group with changing membership merely from general network properties, such as measures of centrality and latent communities? In this paper, we address the problem of identifying social groups from conversation data and present results of an empirical study on identifying groups in a virtual world. Virtual worlds are interesting because group membership is more shaped by common interests and less influenced by cultural and socio-economic factors. Our finding is that a combination of network measures is more predictive of group membership than language cues, and that both types of features can be combined to improve prediction.