Probabilistic Latent Semantic Analysis
Probabilistic Latent Semantic Analysis is a novel statistical technique for the analysis of two-mode and co-occurrence data, which has applications in information retrieval and filtering, natural language processing, machine learning from text, and in related areas. Compared to standard Latent Semantic Analysis which stems from linear algebra and performs a Singular Value Decomposition of co-occurrence tables, the proposed method is based on a mixture decomposition derived from a latent class model. This results in a more principled approach which has a solid foundation in statistics. In order to avoid overfitting, we propose a widely applicable generalization of maximum likelihood model fitting by tempered EM. Our approach yields substantial and consistent improvements over Latent Semantic Analysis in a number of experiments.
Keywords: information retrieval, latent class models, dimension reduction, EM algorithm
PS Link: http://www.icsi.berkeley.edu/~hofmann/Papers/Hofmann-UAI99.ps
PDF Link: /papers/99/p289-hofmann.pdf
AUTHOR = "Thomas Hofmann
TITLE = "Probabilistic Latent Semantic Analysis",
BOOKTITLE = "Proceedings of the Fifteenth Conference Annual Conference on Uncertainty in Artificial Intelligence (UAI-99)",
PUBLISHER = "Morgan Kaufmann",
ADDRESS = "San Francisco, CA",
YEAR = "1999",
PAGES = "289--296"