Generalizing Translation Models in the Probabilistic Relevance Framework

Authors: 
Navid Rekabsaz
Mihai Lupu
Allan Hanbury
Guido Zuccon
Type: 
Speech with proceedings
Proceedings: 
CIKM '16 Proceedings of the 25th ACM International on Conference on Information and Knowledge Management
Publisher: 
ACM
Pages: 
711 - 720
ISBN: 
ISBN: 978-1-4503-4073-1
Year: 
2016
Abstract: 
A recurring question in information retrieval is whether term associations can be properly integrated in traditional information retrieval models while preserving their robustness and effectiveness. In this paper, we revisit a wide spectrum of existing models (Pivoted Document Normalization, BM25, BM25 Verboseness Aware, Multi-Aspect TF, and Language Modelling) by introducing a generalisation of the idea of the translation model. This generalisation is a de facto transformation of the translation models from Language Modelling to the probabilistic models. In doing so, we observe a potential limitation of these generalised translation models: they only affect the term frequency based components of all the models, ignoring changes in document and collection statistics. We correct this limitation by extending the translation models with the 15 statistics of term associations and provide extensive experimental results to demonstrate the benefit of the newly proposed methods. Additionally, we compare the translation models with query expansion methods based on the same term association resources, as well as based on Pseudo-Relevance Feedback (PRF). We observe that translation models always outperform the first, but provide complementary information with the second, such that by using PRF and our translation models together we observe results better than the current state of the art.
TU Focus: 
Computational Science and Engineering
Reference: 

N. Rekabsaz, M. Lupu, A. Hanbury, G. Zuccon:
"Generalizing Translation Models in the Probabilistic Relevance Framework";
Vortrag: 25th ACM International on Conference on Information and Knowledge Management (CIKM), Indianapolis, Indiana, USA; 24.10.2016 - 28.10.2016; in: "CIKM '16 Proceedings of the 25th ACM International on Conference on Information and Knowledge Management", ACM, New York (2016), ISBN: 978-1-4503-4073-1; S. 711 - 720.

Zusätzliche Informationen

Last changed: 
11.12.2016 15:22:23
TU Id: 
253803
Accepted: 
Accepted
Invited: 
Department Focus: 
Business Informatics
Abstract German: 
Author List: 
N. Rekabsaz, M. Lupu, A. Hanbury, G. Zuccon