Learning to detect conversation focus of threaded discussions

Abstract

In this paper we present a novel featureenriched approach that learns to detect the conversation focus of threaded discussions by combining NLP analysis and IR techniques. Using the graph-based algorithm HITS, we integrate different features such as lexical similarity, poster trustworthiness, and speech act analysis of human conversations with featureoriented link generation functions. It is the first quantitative study to analyze human conversation focus in the context of online discussions that takes into account heterogeneous sources of evidence. Experimental results using a threaded discussion corpus from an undergraduate class show that it achieves significant performance improvements compared with the baseline system.

Semi-supervised speech act recognition in emails and forums

“…They defined verb and noun categories for email speech acts and used supervised learning to recognize them. Feng et al (2006) presented a method of detecting conversation focus based on the speech acts of messages in discussion boards. Extending Feng et al (2006)'s work, Ravi and Kim (2007) applied speech act classification to detect unanswered questions.…”

“…Analysis of speech acts for online chat and instant messages and have been studied in computer-mediated communication (CMC) and distance learning (Twitchell et al, 2004;Nastri et al, 2006;Rosé et al, 2008). In natural language processing, Cohen et al (2004) and Feng et al (2006) used speech acts to capture the intentional focus of emails and discussion boards. However, they assume that enough labeled data are available for developing speech act recognition models.…”

Named entity normalization in user generated content

“…Most of the attention has gone to blogs (see [16] for a recent survey on text analytics for blogs). Online discussion fora are more closely related to the data with which we work; recent research includes work on finding authoritative answers in forum threads [9,12], as well as attempts to assess the quality of forum posts [20]. To the best of our knowledge, discussion threads as triggered by news stories of the kind considered here have not been studied before.…”

Towards an on-Line Student Assistant within a CSCL Framework

“…However, one of the most complex problems to assist the participants in on-line mode is to be able of detecting the different categories of messages, and thus defining the kind of assistance that could correspond to that one requested by a particular message. The following approaches, most of them based on the analysis of speech acts derived from dialogs under study, could potentially be adapted and applied to contribute to solve several aspects of this problem: the Cohen's work (Cohen, 2004) aimed at detecting many categories of messages with high precision and moderate recall by using text-classification learning methods; the Feng's work (Feng, 2006) is centered on detecting which messages in a thread contains the focus of the conversation, localizing what messages are related with the subject, recovering past conversations and using them to solve doubts. This work integrates studies of conversational speech acts, an analysis of message values based on poster trustworthiness and an analysis of lexical similarity; in (Ravi & Kim, 2007), speech act classifiers were developed which were able to identify whether a message contains questions or answers; the purpose of a subsequent Kim's work (Kim, 2008) was to develop tools that could automatically assess student participation and promote interactions by sending responses to student messages.…”