Query efficiency prediction for dynamic pruning

Tonellotto, N., Macdonald, C. and Ounis, I. (2011) Query efficiency prediction for dynamic pruning. In: LSDS-IR: Large-Scale and Distributed Systems for Information Retrieval 2011, Glasgow, 28 October 2011, p. 3. (doi: 10.1145/2064730.2064734)

Abstract

Dynamic pruning strategies are effective yet permit efficient retrieval by pruning - i.e. not fully scoring all postings of all documents matching a given query. However, the amount of pruning possible for a query can vary, resulting in queries with similar properties (query length, total numbers of postings) taking different amounts of time to retrieve search results. In this work, we investigate the causes for inefficient queries, identifying reasons such as the balance between informativeness of query terms, and the distribution of retrieval scores within the posting lists. Moreover, we note the advantages in being able to predict the efficiency of a query, and propose various query efficiency predictors. Using 10,000 queries and the TREC ClueWeb09 category B corpus for evaluation, we find that combining predictors using regression can accurately predict query response time.

• Item Type: Conference Proceedings
• Additional Information: <p>LSDS-IR '11 best paper award.</p> <p>"This year's award is given to the paper entitled 'Query efficiency prediction for dynamic pruning' by Nicola Tonellotto, Craig Macdonald, and Iadh Ounis. We congratulate the authors for their great work.</p> <p>The decision is given by taking into account the large amount of discussion this paper generated during the workshop and, more importantly, the positive feedback it received from the reviewers."</p>
• Status: Published
• Refereed: Yes
• Glasgow Author(s) Enlighten ID: Macdonald, Professor Craig and Ounis, Professor Iadh
• Authors: Tonellotto, N., Macdonald, C., and Ounis, I.
• Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
• College/School: College of Science and Engineering > School of Computing Science