Turkish Journal of Electrical Engineering and Computer Sciences
Abstract
A search engine strikes a balance between effectiveness and efficiency to retrieve the best documents in a scalable way. Recent deep learning-based ranker methods are proving to be effective and improving the state-of-the-art in relevancy metrics. However, as opposed to index-based retrieval methods, neural rankers like bidirectional encoder representations from transformers (BERT) do not scale to large datasets. In this article, we propose a query term weighting method that can be used with a standard inverted index without modifying it. Query term weights are learned using relevant and irrelevant document pairs for each query, using a pairwise ranking loss. The learned weights prove to be more effective than term recall which is a probabilistic relevance feedback, previously used for the task. We further show that these weights can be predicted with a BERT regression model and improve the performance of both a BM25 based index and an index already optimized with a term weighting function.
DOI
10.55730/1300-0632.3913
Keywords
Information retrieval, passage ranking, term weighting, pairwise ranking optimization
First Page
1914
Last Page
1930
Recommended Citation
ŞAHİN, Ö, ÇİÇEKLİ, İ, & ERCAN, G (2022). Learning term weights by overfitting pairwise ranking loss. Turkish Journal of Electrical Engineering and Computer Sciences 30 (5): 1914-1930. https://doi.org/10.55730/1300-0632.3913
Included in
Computer Engineering Commons, Computer Sciences Commons, Electrical and Computer Engineering Commons
