Classifying Sentences as Speech Acts in Message Board Posts

Proceedings of the 2011 Conference on Empirical Methods in Natural Language Processing, pages 748–758, Edinburgh, Scotland, UK, July 27–31, 2011. c⃝2011 Association for Computational Linguistics Classifying Sentences as Speech Acts in Message Board Posts Ashequl Qadir and Ellen Riloff School of Computing University of Utah Salt Lake City, UT 84112 {asheq,riloff}@cs.utah.edu Abstract This research studies the text genre of mes- sage board forums, which contain a mix- ture of expository sentences that present fac- tual information and conversational sentences that include communicative acts between the writer and readers. Our goal is to create sentence classifiers that can identify whether a sentence contains a speech act, and can recognize sentences containing four different speech act classes: Commissives, Directives, Expressives, and Representatives. We con- duct experiments using a wide variety of fea- tures, including lexical and syntactic features, speech act word lists from external resources, and domain-specific semantic class features. We evaluate our results on a collection of mes- sage board posts in the domain of veterinary medicine. 1 Introduction In the 1990’s, the natural language processing com- munity shifted much of its attention to corpus-based learning techniques. Since then, most of the text cor- pora that have been annotated and studied are collec- tions of expository text (e.g., news articles, scientific literature, etc.). The intent of expository text is to present or explain information to the reader. In re- cent years, there has been a growing interest in text genres that originate from Web sources, such as we- blogs and social media sites (e.g., tweets). These text genres offer new challenges for NLP, such as the need to handle informal and loosely grammatical text, but they also pose new opportunities to study discourse and pragmatic phenomena that are funda- mentally different in these genres. Message boards are common on the WWW as a forum where people ask questions and post com- ments to members of a community. They are typ- ically devoted to a specific topic or domain, such as finance, genealogy, or Alzheimer’s disease. Some message boards offer the opportunity to pose ques- tions to domain experts, while other communities are open to anyone who has an interest in the topic. From a natural language processing perspective, message board posts are an interesting hybrid text genre because they consist of both expository text and conversational text. Most obviously, the conver- sations appear as a thread, where different people respond to each other’s questions in a sequence of posts. Studying the conversational threads, however, is not the focus of this paper. Our research addresses the issue of conversational pragmatics within indi- vidual message board posts. Most message board posts contain both exposi- tory sentences as well as speech acts. The person posting a message (the “writer”) often engages in speech acts with the readers. The writer may explic- itly greet the readers (“Hi everyone!”), request help from the readers (“Anyone have a suggestion?”), or commit to a future action (“I promise I will report back soon.”). But most posts contain factual …

Content truncated to 500 words