Parsimonious language models for information retrieval

Hiemstra, Djoerd; Robertson, Stephen; Zaragoza, Hugo

doi:10.1145/1008992.1009025

Cited by 159 publications

(203 citation statements)

References 21 publications

Supporting

Mentioning

201

Contrasting

Order By: Relevance

“…We compare the performance of our model against the performance of a standard relevance-based language model (Rel LM). Our implementation of the Rel LM follows the description given in [7]. We also compare performance against a simple no feedback Lucene retrieval baseline [19].…”

Section: Pseudo-relevance Feedbackmentioning

confidence: 99%

See 1 more Smart Citation

A Latent Dirichlet Framework for Relevance Modeling

Ha-Thuc

Srinivasan

2009

Information Retrieval Technology

View full text Add to dashboard Cite

Abstract. Relevance-based language models operate by estimating the probabilities of observing words in documents relevant (or pseudo relevant) to a topic. However, these models assume that if a document is relevant to a topic, then all tokens in the document are relevant to that topic. This could limit model robustness and effectiveness. In this study, we propose a Latent Dirichlet relevance model, which relaxes this assumption. Our approach derives from current research on Latent Dirichlet Allocation (LDA) topic models. LDA has been extensively explored, especially for generating a set of topics from a corpus. A key attraction is that in LDA a document may be about several topics. LDA itself, however, has a limitation that is also addressed in our work. Topics generated by LDA from a corpus are synthetic, i.e., they do not necessarily correspond to topics identified by humans for the same corpus. In contrast, our model explicitly considers the relevance relationships between documents and given topics (queries). Thus unlike standard LDA, our model is directly applicable to goals such as relevance feedback for query modification and text classification, where topics (classes and queries) are provided upfront. Thus although the focus of our paper is on improving relevance-based language models, in effect our approach bridges relevance-based language models and LDA addressing limitations of both. Finally, we propose an idea that takes advantage of "bagof-words" assumption to reduce the complexity of Gibbs sampling based learning algorithm.

show abstract

Section: Pseudo-relevance Feedbackmentioning

confidence: 99%

“…Relevance is a key concept in retrieval theory [7] [14]. Among the formal relevance models that have been proposed, relevance-based language models is perhaps the most popular one [9][10] [11].…”

Section: Introductionmentioning

confidence: 99%

A Latent Dirichlet Framework for Relevance Modeling

Ha-Thuc

Srinivasan

2009

Information Retrieval Technology

View full text Add to dashboard Cite

show abstract

“…In order to provide a better than uniform prior probability of relevance, for a number of experiments the text from search topic descriptions serves as a query to match against the speech transcripts, using a language model [9]. In another version of the system the prior distribution is determined by the number of neighbours in the association matrix for each document, so that a document with many neighbours has higher chance to be displayed.…”

Section: Interactive Experiments Setupmentioning

confidence: 99%

“…Probabilistic approaches from text retrieval (e.g. [9,12] gained less popularity among non-text CBIR researches with some notable exceptions [3,19,16]. One of the reasons lies in the difficulty of translating the lower-level features into probability values.…”

Section: Introductionmentioning

confidence: 99%

Interactive Content-Based Retrieval Using Pre-computed Object-Object Similarities

Boldareva

Hiemstra

2004

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

show abstract

“…The language modeling framework was first introduced by Ponte and Croft [19], followed by many research activities related to this framework since then [1, 3, 4, 8, 10-12, 14-18, 20, 21, 23]. For example, query expansion techniques [3,11,12,17,18,21,23], pseudo-relevance feedback [4,11,12,17,18,21,23], parameter estimation methods [10], multi-word features [20], passage segmentations [16] and time constraints [14] have been proposed to improve the language modeling frameworks. Among them, query expansion with pseudo feedback can increase retrieval performance significantly [11,18,23].…”

Section: Introductionmentioning

confidence: 99%

Enhancing Relevance Models with Adaptive Passage Retrieval

Zhu

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. Passage retrieval and pseudo relevance feedback/query expansion have been reported as two effective means for improving document retrieval in literature. Relevance models, while improving retrieval in most cases, hurts performance on some heterogeneous collections. Previous research has shown that combining passage-level evidence with pseudo relevance feedback brings added benefits. In this paper, we study passage retrieval with relevance models in the language-modeling framework for document retrieval. An adaptive passage retrieval approach is proposed to document ranking based on the best passage of a document given a query. The proposed passage ranking method is applied to two relevance-based language models: the Lavrenko-Croft relevance model and our robust relevance model. Experiments are carried out with three query sets on three different collections from TREC. Our experimental results show that combining adaptive passage retrieval with relevance models (particularly the robust relevance model) consistently outperforms solely applying relevance models on full-length document retrieval.

show abstract

Parsimonious language models for information retrieval

Cited by 159 publications

References 21 publications

A Latent Dirichlet Framework for Relevance Modeling

A Latent Dirichlet Framework for Relevance Modeling

Interactive Content-Based Retrieval Using Pre-computed Object-Object Similarities

Enhancing Relevance Models with Adaptive Passage Retrieval

Contact Info

Product

Resources

About