About learning models with multiple query dependent features

Macdonald, C. , Santos, R. L.T., Ounis, I. and He, B. (2013) About learning models with multiple query dependent features. ACM Transactions on Information Systems, 31(3), (doi: 10.1145/2493175.2493176)

Full text not currently available from Enlighten.

Publisher's URL: http://dx.doi.org/10.1145/2493175.2493176


Several questions remain unanswered by the existing literature concerning the deployment of query dependent features within learning to rank. In this work, we investigate three research questions to empirically ascertain best practices for learning to rank deployments: (i) Previous work in data fusion that pre-dates learning to rank showed that while different retrieval systems could be effectively combined, the combination of multiple models within the same system was not as effective. In contrast, the existing learning to rank datasets (e.g. LETOR), often deploy multiple weighting models as query dependent features within a single system, raising the question as to whether such combination is needed. (ii) Next, we investigate whether the training of weighting model parameters, traditionally required for effective retrieval, is necessary within a learning to rank context. (iii) Finally, we note that existing learning to rank datasets use weighting model features calculated on different fields (e.g. title, content or anchor text), even though such weighting models have been criticised in the literature. Experiments to address these three questions are conducted on Web search datasets, using various weighting models as query dependent, and typical query independent features, which are combined using three learning to rank techniques. In particular, we show and explain why multiple weighting models should be deployed as features. Moreover, we unexpectedly find that training the weighting model’s parameters degrades learned models effectiveness. Finally, we show that computing a weighting model separately for each field is less effective than more theoretically-sound field-based weighting models.

Item Type:Articles
Glasgow Author(s) Enlighten ID:He, Mr Ben and Macdonald, Professor Craig and Ounis, Professor Iadh
Authors: Macdonald, C., Santos, R. L.T., Ounis, I., and He, B.
Subjects:Q Science > QA Mathematics > QA75 Electronic computers. Computer science
College/School:College of Science and Engineering > School of Computing Science
Journal Name:ACM Transactions on Information Systems
Publisher:ACM Press
ISSN (Online):1558-2868

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