Personalised Search Time Prediction using Markov Chains

Tran, V., Maxwell, D., Fuhr, N. and Azzopardi, L. (2017) Personalised Search Time Prediction using Markov Chains. In: ICTIR 2017: The 3rd ACM International Conference on the Theory of Information Retrieval, Amsterdam, The Netherlands, 1-4 Oct 2017, pp. 237-240. ISBN 9781450344906 (doi: 10.1145/3121050.3121085)

149219.pdf - Accepted Version



For improving the effectiveness of Interactive Information Retrieval (IIR), a system should minimise the search time by guiding the user appropriately. As a prerequisite, in any search situation, the system must be able to estimate the time the user will need for finding the next relevant document. In this paper, we show how Markov models derived from search logs can be used for predicting search times, and describe a method for evaluating these predictions. For personalising the predictions based upon a few user events observed, we devise appropriate parameter estimation methods. Our experimental results show that by observing users for only 100 seconds, the personalised predictions are already significantly better than global predictions.

Item Type:Conference Proceedings
Glasgow Author(s) Enlighten ID:MAXWELL, David Martin and Azzopardi, Dr Leif
Authors: Tran, V., Maxwell, D., Fuhr, N., and Azzopardi, L.
College/School:College of Science and Engineering > School of Computing Science
Publisher:ACM Press
Copyright Holders:Copyright © 2017 Association for Computing Machinery
First Published:First published in ICTIR '17 Proceedings of the ACM SIGIR International Conference on Theory of Information Retrieval: 237-240
Publisher Policy:Reproduced in accordance with the publisher copyright policy

University Staff: Request a correction | Enlighten Editors: Update this record