UH-PRHLT at SemEval-2016 Task 3: Combining Lexical and Semantic-based Features for Community Question Answering

Proceedings of SemEval-2016, pages 814–821, San Diego, California, June 16-17, 2016. c⃝2016 Association for Computational Linguistics UH-PRHLT at SemEval-2016 Task 3: Combining Lexical and Semantic-based Features for Community Question Answering Marc Franco-Salvador1, Sudipta Kar2, Thamar Solorio2, and Paolo Rosso1 1 Pattern Recognition and Human Language Technology (PRHLT) research center Universitat Polit`ecnica de Val`encia, Spain 2 Department of Computer Science University of Houston, United States mfranco@prhlt.upv.es, skar3@uh.edu, tsolorio@uh.edu, prosso@prhlt.upv.es Abstract In this work we describe the system built for the three English subtasks of the Se- mEval 2016 Task 3 by the Department of Computer Science of the University of Houston (UH) and the Pattern Recog- nition and Human Language Technology (PRHLT) research center - Universitat Polit`ecnica de Val`encia: UH-PRHLT. Our system represents instances by using both lexical and semantic-based similarity mea- sures between text pairs. Our semantic features include the use of distributed rep- resentations of words, knowledge graphs generated with the BabelNet multilingual semantic network, and the FrameNet lexi- cal database. Experimental results outper- form the random and Google search en- gine baselines in the three English sub- tasks. Our approach obtained the highest results of subtask B compared to the other task participants. 1 Introduction The key role that the Internet plays today for our society benefited the dawn of thousands of new Web social activities. Among those, fo- rums emerged with special relevance following the paradigm of the Community Question An- swering (CQA). These type of social networks allow people to post a question to other users of that community. The usage is simple, without much restrictions, and infrequently moderated. The popularity of CQA is a strong indicator that users receive some good and valuable answers. However, there are several issues related to that type of community. First is the large amount of answers received that makes it difficult and time-consuming for users to search and distin- guish the good ones. This is exacerbated with the amount of noise that these questions con- tain. It is not uncommon to have wrong or mis- guiding answers that produce more unrelated answers, discussions and sub-threads. Finally, there is a lot of redundancy, questions may be repeated or closely related to previously asked questions. Details of the SemEval 2016 Task 3 on CQA can be found in the overview paper (Nakov et al., 2016). In this work we evaluate the three English-related Task 3 subtasks on CQA. We first represent each instance to rank — ques- tion versus (vs.) comments, question vs. related questions, or question vs. comments of related questions — with a set of similarities computed at two different levels: lexical and semantic. This representation allows us to estimate the re- latedness between text pairs in terms of what is explicitly stated and what it means. Our lex- ical similarities employ representations such as word and character n-grams, and bag-of-words (BOW). The semantic similarities include the use of distributed word bidirectional alignments, distributed representations of text, knowledge graphs, and frames from the FrameNet lexical database (Baker et al., …

Content truncated to 500 words