Premium
Improving answer selection with global features
Author(s) -
Gu Shengwei,
Luo Xiangfeng,
Wang Hao,
Huang Jing,
Wei Qin,
Huang Subin
Publication year - 2021
Publication title -
expert systems
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.365
H-Index - 38
eISSN - 1468-0394
pISSN - 0266-4720
DOI - 10.1111/exsy.12603
Subject(s) - computer science , selection (genetic algorithm) , artificial intelligence , task (project management) , question answering , machine learning , rank (graph theory) , deep learning , deep neural networks , feature selection , mean reciprocal rank , feature (linguistics) , point (geometry) , reciprocal , information retrieval , linguistics , philosophy , geometry , mathematics , management , combinatorics , economics
Given a question and its answer candidates (named QA corpus), answer selection is the task of identifying the most relevant answers to the question. Answer selection is widely used in question answering, web search, and so on. Current deep neural network models primarily utilize local features extracted from input question‐answer pairs (QA pairs). However, the global features contained in QA corpora are under‐utilized, and we argue that these global features substantially contribute to the answer selection task. To verify this point of view, we propose a novel model that combines local and global features for answer selection. In our model, two different global feature extractors are employed to extract statistical global features and deep global features from a QA corpus, respectively. Furthermore, we investigate the integration of these global features with local features in various experimental settings: statistical global features, deep global features, and a combination of statistical and deep global features. Our experimental results show that the global features are effective for answer selection. Our model obtains new state‐of‐the‐art results on two public answer selection datasets and performs especially well on YahooCQA, where it achieves 9.2 and 6% higher precision@1 (P@1) and mean reciprocal rank (MRR) scores than previously published models.