Predicting query performance

By Steve Cronen-Townsend, Yun Zhou, W. Bruce Croft. Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, 2002

826 citations recorded by the source corpus.

Subfields: Artificial Intelligence

View Paper: https://www.semanticscholar.org/paper/0c7a7e6a0aa405f6cc6b686915731865222b6e4b

How researchers see this paper

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

In Artificial Intelligence

Citations in the corpus

Ask the Right Questions: Active Question Reformulation with Reinforcement Learning (2017, neutral): “er query, where the document is the context from which the answer is selected, as a measure of how informative a term is.4 As another measure of query performance, we also compute Query Clarity (QC) (Cronen-Townsend et al., 2002).5 Figure 2 summarizes statistics of the questions and rewrites. We ﬁrs”
Information Retrieval Evaluation (2011, background): “A fruitful area for this investigation would be the causes of huge variations in system performance across topics, and there have been multiple papers trying to find ways of predicting performance on a given topic (for example [40, 62] and another lecture in this series “Estimating the Query Difficu”
Learning to model relatedness for news recommendation (2011, methodology): “), we compute a relative entropy or KL-divergence, similarly to [9].”
Book Review: Statistical Language Models for Information Retrieval by ChengXiang Zhai (2008, background): “© ChengXiang Zhai, 2006 81 Predicting Query Difficulty [Cronen-Townsend et al. 02] • Observations: – Discriminative queries tend to be easier – Comparison of the query model and the collection model can indicate how discriminative a query is • Method: – Define “query clarity” as the KL-divergence be”
Novelty and diversity in information retrieval evaluation (2008, background): “” A number of researchers have investigated the relationship between query ambiguity and query difficulty [4, 14, 19].”
Personalized query expansion for the web (2007, methodology): “• The Query Clarity [7] seems to be the best, as well as the most applied technique so far.”
Modeling and solving term mismatch for full-text retrieval (2012, methodology): “2002) can be used as a measure of word abstractness, where each word is treated as an individual query and query clarity is measured according to (Cronen-Townsend et al. 2002).”
Linear feature-based models for information retrieval (2007, background): “Examples of these features are PageRank [2], URL depth [14], document quality [35], readability [31], sentiment [25], and query clarity [7].”
Novel Visual and Statistical Image Features for Microblogs News Verification (2017, background): “Based on their success in image retrieval, these visual features, such as clarity score [26], [27], coherence score [28], [29] and diversity score [30]–[32], are expected to be effective in news verification task.”
Information Retrieval: A Biomedical and Health Perspective (2020, result): “also showed that younger reviewers and those who had training in epidemiology or statistics produced better reviews, though their study found in general that reviewer characteristics could explain only 8% of the variation in review quality [179].”
Query Expansion Techniques for Information Retrieval: a Survey (2017, neutral): “Here, the number of expansion terms depend on the ambiguity of the initial query in the web or the user log; the ambiguity is determined by the clarity score [72].”
A comparison of user and system query performance predictions (2010, methodology): “Clarity Score [8] is based on the intuition that the top m ranked documents of an unambiguous query will be topically cohesive and terms particular to the topic will appear with high frequency.”
MIMICS: A Large-Scale Data Collection for Search Clarification (2020, background): “A major task in search clarification is deciding whether to ask clarification, especially in search systems with limited-bandwidth interfaces. This problem can be cast to query performance prediction [6, 10]. In other words, clarification can be asked when the predicted performance for the given que”
To personalize or not to personalize: modeling queries with variation in user intent (2008, methodology): “[4], is a measure of the quality of the results returned for a query that does not require the query to have been seen by a search engine before.”