Topic-Based Clusters in Egocentric Networks on Facebook
Keywords:homophily, egocentric network, Facebook, topic, topic interests, social circles
Homophily suggests that people tend to befriend others with shared traits, such as similar topical interests or overlapping social circles. We study how people communicate online in term of conversation topics from an egocentric viewpoint using a dataset from Facebook. We find that friends who favor similar topics form topic-based clusters; these clusters have dense connectivities, large growth rates, and little overlap.