Probabilistic Latent Semantic Analysis

DMCA

Probabilistic Latent Semantic Analysis (1999)

Cached

Download Links

Other Repositories/Bibliography

CiteULike
DBLP

by Thomas Hofmann

Venue:	In Proc. of Uncertainty in Artificial Intelligence, UAI’99
Citations:	754 - 9 self

BibTeX

@INPROCEEDINGS{Hofmann99probabilisticlatent,
    author = {Thomas Hofmann},
    title = {Probabilistic Latent Semantic Analysis},
    booktitle = {In Proc. of Uncertainty in Artificial Intelligence, UAI’99},
    year = {1999},
    pages = {289--296}
}

OpenURL

Abstract

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.

Keyphrases

probabilistic latent semantic analysis latent semantic analysis singular value decomposition latent class model related area novel statistical technique approach yield co-occurrence data natural language processing information retrieval applicable generalization tempered em consistent improvement machine learning principled approach mixture decomposition co-occurrence table solid foundation linear algebra maximum likelihood model fitting

DMCA