Feature-based question routing in community question answering platforms

• Published in: Information sciences Vol. 608; pp. 696 - 717
• Main Authors: Sorkhani, Soroosh, Etemadi, Roohollah, Bigdeli, Amin, Zihayat, Morteza, Bagheri, Ebrahim
• Format: Journal Article
• Language: English
• Published: Elsevier Inc 01.08.2022
• Subjects: Embedding; Expert finding; Learning to rank; Question routing; Question routing; Embedding; Learning to rank; Expert finding
• ISSN: 0020-0255; 1872-6291
• DOI: 10.1016/j.ins.2022.06.072

•Proposed a Novel learn to rank model with explainable results for question routing.•Provided a systematic classification of and the introduction of 74 features that can be effective for the task of question routing.•Offered in-depth analysis on the impact of provided features and feature categories for question routing.•Adopted and reported on the Gini score to enhance interpretability of the results.•Compare the performance of the proposed model against strong neural methods for question routing. Community question answering (CQA) platforms are receiving increased attention and are becoming an indispensable source of information in different domains ranging from board games to physics. The success of these platforms dependent on how efficiently new questions are assigned to community experts, known ascalled question routing. In this paper, we address the problem of question routing by adopting a learning to rank approach over five CQA websites in the context of which we introduce 74 features and systematically classify them into content-based and social-based categories. Our extensive experiments on datasets from five real online question answering websites indicate that content-based features related to tags and topics as well as social features that are related to user characteristics and user temporality are effective for question routing. Our work shows the ability to improve performance compared to the state-of-the-art neural matchmaking methods that lack the interpretability offered by our work. The improvement can be as high as on average 2.47% and 1.10% in terms of common ranking metrics, Normalized Discounted Cumulative Gain (NDCG) and Mean Average Precision (MAP) respectively, compared to our best baselines.