Abstract

Probabilistic Latent Semantic Indexing is a novel approach to automated document indexing which is based on a statistical latent class model for factor analysis of count data. Fitted from a training corpus of text documents by a generalization of the Expectation Maximization algorithm, the utilized model is able to deal with domain-specific synonymy as well as with polysemous words. In contrast to standard Latent Semantic Indexing (LSI) by Singular Value Decomposition, the probabilistic variant has a solid statistical foundation and defines a proper generative data model. Retrieval experiments on a number of test collections indicate substantial performance gains over direct term matching methodsaswell as over LSI. In particular, the combination of models with different dimensionalities has proven to be advantageous.

Citations

6231	Maximum likelihood from incomplete data via the EM algorithm - Dempster, Laird - 1977
2699	MJ: Introduction to Modern Information Retrieval - Salton, McGill - 1986
2168	R.A.: Indexing by latent semantic analysis - Deerwester, Dumais, et al. - 1990
612	A view of the EM algorithm that justifies incremental, sparse, and other variants - Neal, Hinton - 1998
595	Finite Mixture Models - McLachlan, Peel - 1997
477	Distributional clustering of english words - Pereira, Tishby, et al. - 1993
375	Probabilistic latent semantic analysis - Hofmann - 1999
128	Latent class models for collaborative filtering - Hofmann, Puzicha - 1999
89	Unsupervised learning from dyadic data - Hofmann, Puzicha - 1998
72	A deterministic annealing approach to clustering - Rose, Gurewitz, et al. - 1990
46	Latent Semantic Indexing (LSI): TREC-3 report," in Overview of the Third Text REtrieval Conference - Dumais - 1994
46	G.(1998): Predicting the Performance of Linearly Combined IR Systems - Vogt, Cottrell
37	Differential geometry and statistics - Murray, Rice - 1993
35	Topic-based language models using EM - Gildea, Hofmann - 1999
26	A view of the EM algorithm that justi es incremental, sparse, and other variants - Neal, Hinton - 1998
7	erential Geometry and Statistics - Murray, Rice, et al. - 1993
6	Aggregate and mixed{ order Markov models for statistical language processing - Saul, Pereira - 1997
2	Latent class models for collaborative ltering - Hofmann - 1999