摘要：Summary With the growing importance for analysis of the textual similarity and between the user contents, we have highlighted textual similarity between various people in a social network. Words used in social sites are used for finding textual similarity. On the basis of the common words used in social networks, we have formulated a metric. The data has been extracted from social networking sites and then it is processed for generating the metrics. We compare simple k-means and spectral k-means algorithms for finding textual similarity. We have used WordNet to groups words together based on their meanings.
关键词：Social network analysis; Data mining; k-means; Spectral clustering; Twitter;
