What question did this study set out to answer?

To develop a probabilistic method for automated document indexing that improves semantic understanding of text.

August 1, 1999Open Access

Probabilistic latent semantic indexing

View Full Paper

Ask AI

Bookmark

Authors

Randomized trial demonstrates improved indexing performance in text documents, suggesting a robust model for retrieval.

To develop a probabilistic method for automated document indexing that improves semantic understanding of text.
Utilized a statistical latent class model for factor analysis of count data.
Applied the Expectation Maximization algorithm to train on a corpus of text documents.
Conducted retrieval experiments on multiple test collections.
Achieved significant performance gains in document retrieval compared to direct term matching methods.
Demonstrated superior results over standard Latent Semantic Indexing (LSI).
Combination of models with different dimensionalities enhanced indexing effectiveness.

Thomas Hofmann (1999) studied this question.

View Full Paper

Ask AI

Bookmark