Title:Large scale link based latent Dirichlet allocation for web document classification

Abstract:In this paper we demonstrate the applicability of latent Dirichlet allocation (LDA) for classifying large Web document collections. One of our main results is a novel influence model that gives a fully generative model of the document content taking linkage into account. In our setup, topics propagate along links in such a way that linked documents directly influence the words in the linking document. As another main contribution we develop LDA specific boosting of Gibbs samplers resulting in a significant speedup in our experiments. The inferred LDA model can be applied for classification as dimensionality reduction similarly to latent semantic indexing. In addition, the model yields link weights that can be applied in algorithms to process the Web graph; as an example we deploy LDA link weights in stacked graphical learning. By using Weka's BayesNet classifier, in terms of the AUC of classification, we achieve 4% improvement over plain LDA with BayesNet and 18% over this http URL with SVM. Our Gibbs sampling strategies yield about 5-10 times speedup with less than 1% decrease in accuracy in terms of likelihood and AUC of classification.