APPLICATION OF TOPIC MODELING IN DOCUMENT CLUSTERING

Abstract

This paper is an experimental study aimed at evaluating the effectiveness of distance measures when using the combined model of LDA and K-means for clustering document. Our experimental results indicate that the probabilistic-based distance measures are better than the vector based distance measures including Euclidian when it comes to cluster a set of documents in the topic space. By choosing the probabilistic-based distance measures, K-means combined to the results of the Latent Dirichlet Allocation allows us to have better results than the LDA + Naïve and Vector Space Model