What question did this study set out to answer?

The central aim is to explore the effect of smoothing methods on language models in information retrieval and how they impact retrieval performance.

A study of smoothing methods for language models applied to information retrieval

View Full Paper

Ask AI

Bookmark

Authors

Investigates language model smoothing to improve retrieval performance for different query types.

The central aim is to explore the effect of smoothing methods on language models in information retrieval and how they impact retrieval performance.
Examined several popular smoothing methods across different test collections.
Analyzed the sensitivity of retrieval performance to smoothing parameters based on query types.
Proposed a two-stage smoothing strategy with automated parameter estimation.
Retrieval performance was found to be sensitive to smoothing parameters, especially for verbose queries.
Verbose queries required more aggressive smoothing to optimize performance.
The two-stage smoothing method outperformed traditional single smoothing methods in retrieval accuracy.

Zhai et al. (2004) studied this question.

View Full Paper

Ask AI

Bookmark