Probabilistic Latent Semantic Analysis
Thomas Hofmann · arXiv (Cornell University) · 2013
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.
Cited by 2,047 works. doi:10.48550/arxiv.1301.6705 Open-access PDF
Authors
References
- Maximum Likelihood from Incomplete Data Via the EM Algorithm
- Indexing by latent semantic analysis
- Introduction to Modern Information Retrieval
- A solution to Plato's problem: The latent semantic analysis theory of acquisition, induction, and representation of knowledge.
- Probabilistic Latent Semantic Indexing
- Distributional clustering of English words
- Unsupervised Learning by Probabilistic Latent Semantic Analysis
- Personalized information delivery
- A deterministic annealing approach to clustering
- Aggregate and mixed-order Markov models for statistical language processing