Automatically generating effective search queries directly from community question-answering questions for finding related questions
作者：
Highlights：
• We produce search strings from cQA questions for retrieving related questions.
• Search strings are generated via examining lexicalized dependency paths.
• We tested several Learning to Rank methods.
• Neural Net based methods are likely to perform the best.
• NER and POS taggers as well as WordNet provide effective features.
摘要
•We produce search strings from cQA questions for retrieving related questions.•Search strings are generated via examining lexicalized dependency paths.•We tested several Learning to Rank methods.•Neural Net based methods are likely to perform the best.•NER and POS taggers as well as WordNet provide effective features.
论文关键词：Real-time intelligent automation,Expert systems,Knowledge bases,Knowledge processing,Natural language processing,Community question answering,Question analysis
论文评审过程：Received 24 May 2016, Revised 7 December 2016, Accepted 25 January 2017, Available online 1 February 2017, Version of Record 8 February 2017.
论文官网地址：https://doi.org/10.1016/j.eswa.2017.01.041
