Selecting good expansion terms for pseudo-relevance feedback

By Guihong Cao, Jian-Yun Nie, Jianfeng Gao, Stephen E. Robertson. Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, 2008

477 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Information Systems

View Paper: https://www.semanticscholar.org/paper/bffc8ce2d9f4b2e5cd4573f7eb87c382b015dcd3

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Information Systems

Citations in the corpus

The Simplest Thing That Can Possibly Work: Pseudo-Relevance Feedback Using Text Classification (2019, background): “Note that while there is literature on how to select good expansion terms using supervised machine learning techniques [4], my approach is completely unsupervised .”
Demonstrate-Search-Predict: Composing retrieval and language models for knowledge-intensive NLP (2022, background): “1 def conversational_multihop_search(x): Similar approaches can be used for correcting spelling mistakes or implementing pseudo-relevance feedback (Cao et al., 2008; Wang et al., 2022a), in which retrieved passages are used to inform a better search query, though this has not been attempted with…”
Word Embedding Causes Topic Shifting; Exploit Global Context! (2017, background): “As expected, the percentage of the good terms is higher for the larger threshold, however—similar to the observation on pseudo-relevance feedback [2]—most of the expanded terms (58% to 81%) have no significant effect on performance.”
Approximation of Response Knowledge Retrieval in Knowledge-grounded Dialogue Generation (2020, methodology): “We also use the post related results to create an extended representation of the post, similar to the pseudo-relevancefeedback in query-based search (Cao et al., 2008).”
Query dependent pseudo-relevance feedback based on wikipedia (2009, methodology): “This method is inspired by the work of Cao et al. [3], where they suggest that “good” terms should be identiﬁed directly according to their possible impact on retrieval eﬀectiveness.”
Response Ranking with Deep Matching Networks and External Knowledge in Information-seeking Conversation Systems (2018, background): “Pseudo-relevance feedback (PRF) has been proven effective in improving the performance of many retrieval models [3, 12, 17, 25, 46, 47].”
Answering Complex Open-Domain Questions with Multi-Hop Dense Retrieval (2020, methodology): “In particular, our system is similar in spirit to pseudo-relevance feedback techniques (Croft & Harper, 1979; Cao et al., 2008; Lv & Zhai, 2010), where no additional user interaction is required at the query reformulation stage.”
A Survey of Automatic Query Expansion in Information Retrieval (2012, methodology): “2005] Corpus Web table and form mining Attribute-value pairs Association rules Boolean query [Collins-Thompson and Callan 2005] Stemmer + Wordnet + corpus + top-ranked docs Probabilisticterm association network Single words Markov chain Structured query [Song et al. 2006] Top-ranked docs +corpus K”
CrisisLex: A Lexicon for Collecting and Filtering Microblogged Communications in Crises (2014, background): “Some of the selected terms might be harmful (Cao et al. 2008).”
Multi-step Entity-centric Information Retrieval for Multi-Hop Question Answering (2019, unknown): “, 2018) and for better query reformulation (Cao et al., 2008; Dalton et al., 2014; Dietz and Verga, 2014).”
Entity-Duet Neural Ranking: Understanding the Role of Knowledge Graph Semantics in Neural Information Retrieval (2018, unknown): “They utilize knowledge as a kind of pseudo relevance feedback corpus (Cao et al., 2008) or weight words to better represent query according to well-formed entity descriptions.”
Query Expansion with Freebase (2015, methodology): “However, the effectiveness on the ClueWeb09 and ClueWeb12 web corpora is mixed when compared with strong, state-of-the-art query expansion baselines.”
BERT-QE: Contextualized Query Expansion for Document Re-ranking (2020, methodology): “Cao et al. (2008) perform query expansion by using classification models to select expansion terms.”
Query Expansion Techniques for Information Retrieval: a Survey (2017, unknown): “Reference [55] uses a classifier to recognize the relevant or irrelevant expansion terms.”
Placing flickr photos on a map (2009, methodology): “…prelim-inary knowledge about tags into our models using a simple boosting approach, similar to the one recently used by Cao et al. to boost expansion terms [5]: where P ( Loc | t ) is a probability of the tag t to be location-speciﬁc, β is a boosting coeﬃcient and Z is a normalisation coeﬃcient.”
A Neural Autoencoder Approach for Document Ranking and Query Refinement in Pharmacogenomic Information Retrieval (2018, methodology): “In our approach we follow Cao et al. (2008), in which the ranked documents D q are used as pseudo relevance feedback.”
Clickthrough-based translation models for web search: from word models to phrase models (2010, methodology): “A variety of heuristic and statistical techniques are used to select and (re)weight the expansion terms [30, 35, 11, 5].”
Task-Oriented Query Reformulation with Reinforcement Learning (2017, methodology): “Supervised Learning (SL): Here we detail a deep learning-based variant of the method proposed by Cao et al. (2008).”