On Improving Pseudo-Relevance Feedback Using Pseudo-Irrelevant Documents

Proceedings of the 35th International ACM SIGIR Conference on Research and Development in Information Retrieval

Zheng

2012

Rocchio's relevance feedback model is a classic query expansion method and it has been shown to be effective in boosting information retrieval performance. The selection of expansion terms in this method, however, does not take into account the relationship between the candidate terms and the query terms (e.g., term proximity). Intuitively, the proximity between candidate expansion terms and query terms can be exploited in the process of query expansion, since terms closer to query terms are more likely to be related to the query topic.In this paper, we study how to incorporate proximity information into the Rocchio's model, and propose a proximitybased Rocchio's model, called PRoc, with three variants. In our PRoc models, a new concept (proximity-based term frequency, ptf ) is introduced to model the proximity information in the pseudo relevant documents, which is then used in three kinds of proximity measures. Experimental results on TREC collections show that our proposed PRoc models are effective and generally superior to the state-of-the-art relevance feedback models with optimal parameters. A direct comparison with positional relevance model (PRM) on the GOV2 collection also indicates our proposed model is at least competitive to the most recent progress.

Section: Introductionmentioning

confidence: 98%

Proximity-based rocchio's model for pseudo relevance

Miao

Proceedings of the 35th International ACM SIGIR Conference on Research and Development in Information Retrieval

Zheng

2012

“…With the refined query, usually better retrieval performance can be expected. PRF has been shown to be effective with various retrieval models [6,10,14,17,29,32,34,35,42,44,46,43,25]. There are a large number of studies on the topic of PRF.…”

Section: Related Workmentioning

confidence: 98%

“…It assumes that top-ranked documents in the first-pass retrieval are relevant, and then used as feedback documents in order to refine the representation of original queries by adding potentially related terms or adjusting the weights of query terms. PRF has been shown to be effective in improving IR performance [6,10,14,17,29,32,34,35,42,44,46] in a number of IR tasks.…”

Section: Introductionmentioning

confidence: 99%

A simple term frequency transformation model for effective pseudo relevance feedback

Zheng

Proceedings of the 37th International ACM SIGIR Conference on Research &Amp; Development in Information Retrieval

2014

Pseudo Relevance Feedback is an effective technique to improve the performance of ad-hoc information retrieval. Traditionally, the expansion terms are extracted either according to the term distributions in the feedback documents; or according to both the term distributions in the feedback documents and in the whole document collection. However, most of the existing models employ a single term frequency normalization mechanism or criteria that cannot take into account various aspects of a term's saliency in the feedback documents. In this paper, we propose a simple and heuristic, but effective model, in which three term frequency transformation techniques are integrated to capture the saliency of a candidate term associated with the original query terms in the feedback documents. Through evaluations and comparisons on six TREC collections, we show that our proposed model is effective and generally superior to the recent progress of relevance feedback models.

“…Two categories of studies have been conducted on the topic of PRF. For example, Raman et al (2010) utilized irrelevant feedback documents to extract specific expansion terms from top-ranked documents and showed that good expansion terms are those that discriminate the top-ranked documents. For example, Raman et al (2010) utilized irrelevant feedback documents to extract specific expansion terms from top-ranked documents and showed that good expansion terms are those that discriminate the top-ranked documents.…”

Section: Related Workmentioning

confidence: 99%

“…One category has mainly concentrated on investigating the different impacts of the relevant and irrelevant feedback documents or how to reduce the negative impact of the irrelevant documents (for instance, Mbarek et al (2017); Raman, Udupa, Bhattacharyya, & Bhole, 2010;Singhal, Mitra, & Buckley, 1997;Wang, Fang, & Zhai, 2008;Basile, Caputo, & Semeraro, 2011). For example, Raman et al (2010) utilized irrelevant feedback documents to extract specific expansion terms from top-ranked documents and showed that good expansion terms are those that discriminate the top-ranked documents. The other category involves proposed methods of improving QE processing (for instance, Carpineto et al, 2001;Collins-Thompson, 2009;Greenberg, 2001b;Bai, Song, Bruza, Nie, & Cao, 2005).…”

Section: Related Workmentioning

confidence: 99%

A simple kernel co‐occurrence‐based enhancement for pseudo‐relevance feedback

Pan

Asso for Info Science & Tech

et al. 2019

Pseudo-relevance feedback is a well-studied query expansion technique in which it is assumed that the topranked documents in an initial set of retrieval results are relevant and expansion terms are then extracted from those documents. When selecting expansion terms, most traditional models do not simultaneously consider term frequency and the co-occurrence relationships between candidate terms and query terms. Intuitively, however, a term that has a higher co-occurrence with a query term is more likely to be related to the query topic. In this article, we propose a kernel co-occurrence-based framework to enhance retrieval performance by integrating term co-occurrence information into the Rocchio model and a relevance language model (RM3). Specifically, a kernel co-occurrence-based Rocchio method (KRoc) and a kernel co-occurrence-based RM3 method (KRM3) are proposed. In our framework, co-occurrence information is incorporated into both the factor of the term discrimination power and the factor of the within-document term weight to boost retrieval performance. The results of a series of experiments show that our proposed methods significantly outperform the corresponding strong baselines over all data sets in terms of the mean average precision and over most data sets in terms of P@10. A direct comparison of standard Text Retrieval Conference data sets indicates that our proposed methods are at least comparable to state-of-the-art approaches.