|
Abstract:
|
Latent Dirichlet Allocation (LDA)is an important tool for topic modeling, and it lends itself to clustering documents, movies, and restaurants. This talk describes how LDA improves cluster discovery in a network of political blogs. Specifically, bloggers who are interested in the same topic are more likely to link to each other, and thus the LDA topic analysis improves community detection. The reverse is also true: if two bloggers are in the same cluster, then they are likely to be interested in the same topic, so the clique structure informs the topic discovery.
|
