Download:
|
by Hugo Zaragoza, Djoerd Hiemstra
In Proc. SIGIR 2003
http://research.microsoft.com/~hugoz/pubs/ps/hugoz_sigir03.ps.gz
Add To MetaCart
Abstract:
We propose a Bayesian extension to the ad-hoc Language Model. Many smoothed estimators used for the multinomial query model in ad-hoc Language Models (including Laplace and Bayes-smoothing) are approximations to the Bayesian predictive distribution. In this paper we derive the full predictive distribution in a form amenable to implementation by classical IR models, and then compare it to other currently used estimators. In our experiments the proposed model outperforms Bayes-smoothing, and its combination with linear interpolation smoothing outperforms all other estimators. Categories and Subject Descriptors
Citations
|
500
|
Bayesian Data Analysis
– Gelman, Carlin, et al.
- 1995
|
|
418
|
A language modeling approach to information retrieval
– Ponte, Croft
- 1998
|
|
335
|
An empirical study of smoothing techniques for language modeling
– Chen, Goodman
- 1996
|
|
231
|
A study of smoothing methods for language models applied to ad hoc information retrieval
– Zhai, Lafferty
- 2001
|
|
151
|
Information Retrieval as statistical translation
– Berger, Lafferty
- 1999
|
|
124
|
Relevance-based language models
– Lavrenko, Croft
- 2001
|
|
109
|
Language Modeling for Information Retrieval
– Croft, Lafferty
- 2003
|
|
86
|
Twenty-one at TREC-7: Ad-hoc and cross-language track
– Hiemstra, Kraaij
- 1998
|
|
85
|
Spline Models for Observational Data. Volume 59
– Wahba
- 1990
|
|
46
|
A hierarchical dirichlet language model
– MacKay, Peto
- 1995
|
|
29
|
BBN at TREC7: Using hidden markov models for information retrieval
– Miller, Leek, et al.
- 1998
|
|
20
|
Bayesian data analysis. Chapman and Hall/CRC
– Gelman, Carlin, et al.
- 2004
|
|
3
|
Language models and probability of relevance
– Robertson, Hiemstra
- 2001
|
|
2
|
A hierarchical dirichlet language model
– McKay, Peto
- 1995
|