<?xml version="1.0" standalone="yes"?>
<Paper uid="N06-1027">
  <Title>Learning to Detect Conversation Focus of Threaded Discussions</Title>
  <Section position="3" start_page="208" end_page="209" type="relat">
    <SectionTitle>
2 Related Work
</SectionTitle>
    <Paragraph position="0"> Human conversation refers to situations where two or more participants freely alternate in speaking (Levinson, 1983). What makes threaded discussions unique is that users participate asynchronously and in writing. We model human conversation as a set of messages in a threaded discussion using a graph-based algorithm.</Paragraph>
    <Paragraph position="1"> Graph-based algorithms are widely applied in link analysis and for web searching in the IR community. Two of the most prominent algorithms are Page-Rank (Brin and Page, 1998) and the HITS algorithm (Kleinberg, 1999). Although they were initially proposed for analyzing web pages, they proved useful for investigating and ranking structured objects. Inspired by the idea of graph based algorithms to collectively rank and select the best candidate, research efforts in the natural language community have applied graph-based approaches on keyword selection (Mihalcea and Tarau, 2004), text summarization (Erkan and Radev, 2004; Mihalcea, 2004), word sense disambiguation (Mihalcea et al., 2004; Mihalcea, 2005), sentiment analysis (Pang and Lee, 2004), and sentence retrieval for question answering (Otterbacher et al., 2005). However, until now there has not been any published work on its application to human conversation analysis specifically in the format of threaded discussions. In this paper, we focus on using HITS to detect conversation focus of threaded discussions.</Paragraph>
    <Paragraph position="2"> Rhetorical Structure Theory (Mann and Thomson, 1988) based discourse processing has attracted much attention with successful applications in sentence compression and summarization. Most of the current work on discourse processing focuses on sentence-level text organization (Soricut and Marcu, 2003) or the intermediate step (Sporleder and Lapata, 2005). Analyzing and utilizing discourse information at a higher level, e.g., at the paragraph level, still remains a challenge to the natural language community. In our work, we utilize the discourse information at a message level. Zhou and Hovy (2005) proposed summarizing threaded discussions in a similar fashion to multi-document summarization; but then their work does not take into account the relative importance of different messages in a thread. Marom and Zukerman (2005) generated help-desk responses using clustering techniques, but their corpus is composed of only two-party, two-turn, conversation pairs, which precludes the need to determine relative importance as in a multi-ply conversation.</Paragraph>
    <Paragraph position="3"> In our previous work (Feng et al., 2006), we implemented a discussion-bot to automatically answer student queries in a threaded discussion but extract potential answers (the most informative message) using a rule-based traverse algorithm that is not optimal for selecting a best answer; thus, the result may contain redundant or incorrect information. We argue that pragmatic knowledge like speech acts is important in conversation focus analysis. However, estimated speech act labeling between messages is not sufficient for detecting  human conversation focus without considering other features like author information. Carvalho and Cohen (2005) describe a dependency-network based collective classification method to classify email speech acts. Our work on conversation focus detection can be viewed as an immediate step following automatic speech act labeling on discussion threads using similar collective classification approaches. null We next discuss our approach to detect conversation focus using the graph-based algorithm HITS by taking into account heterogeneous features.</Paragraph>
  </Section>
class="xml-element"></Paper>