Learning and optimization of an aspect hidden markov model for query language model generation.
The Relevance Model (RM) incorporates pseudo relevance feedback to derive query language model and has shown a good performance. Generally, it is based on uni-gram models of individual feedback documents from which query terms are sampled independently. In this paper, we present a new method to build the query model with latent state machine (LSM) which captures the inherent term dependencies within the query and the term dependencies between query and documents. Our method firstly splits the query into subsets of query terms (i.e., not only single terms, but different combinations of multiple query terms). Secondly, these query term combinations are then considered as weighted latent states of a hidden Markov Model to derive a new query model from the pseudo relevant documents. Thirdly, our method integrates the Aspect Model (AM) with the EM algorithm to estimate the parameters involved in the model. Specifically, the pseudo relevant documents are segmented into chunks, and different chunks are associated with different weights in relation to a latent state. Our approach is empirically evaluated on three TREC collections, and demonstrates statistically significant improvements over a baseline language model and the Relevance Model.
HUANG, Q., SONG, D., RUGER, S. and BRUZA, P. 2007. Learning and optimization of an aspect hidden markov model for query language model generation. In Dominich, S. and Kiss, F. (eds.) Studies in theory of information retrieval: proceedings of the 1st Association of Computing Machinery Special Interest Group on Information Retrieval (ACM SIGIR) international conference on the theory of information retrieval (ICTIR'07), 18-20 October 2007, Budapest, Hungary. Budapest: Foundation for Information Society (INFOTA), pages 157-164.
|Conference Name||1st Association of Computing Machinery Special Interest Group on Information Retrieval (ACM SIGIR) international conference on the theory of information retrieval (ICTIR'07)|
|Conference Location||Budapest, Hungary|
|Start Date||Oct 18, 2007|
|End Date||Oct 20, 2007|
|Acceptance Date||Oct 18, 2007|
|Publication Date||Dec 31, 2007|
|Deposit Date||Aug 20, 2009|
|Publicly Available Date||Aug 20, 2009|
|Publisher||INFOTA Foundation for Information Society|
|Keywords||Aspect model; Latent variable model; Segmentation; Information retrieval|
HUANG 2007 Learning and optimization
Publisher Licence URL
You might also like
Predicting emotional reaction in social networks.
Early fusion and query modification in their dual late fusion forms.
You have e-mail, what happens next? Tracking the eyes for genre.