<?xml version="1.0" standalone="yes"?>
<Paper uid="N06-1027">
  <Title>Learning to Detect Conversation Focus of Threaded Discussions</Title>
  <Section position="2" start_page="0" end_page="208" type="intro">
    <SectionTitle>
1 Introduction
</SectionTitle>
    <Paragraph position="0"> Threaded discussion is popular in virtual cyber communities and has applications in areas such as customer support, community development, interactive reporting (blogging) and education. Discussion threads can be considered a special case of human conversation, and since we have huge repositories of such discussion, automatic and/or semi-automatic analysis would greatly improve the navigation and processing of the information.</Paragraph>
    <Paragraph position="1"> A discussion thread consists of a set of messages arranged in chronological order. One of the main challenges in the Question Answering domain is how to extract the most informative or important message in the sequence for the purpose of answering the initial question, which we refer to as the conversation focus in this paper. For example, people may repeatedly discuss similar questions in a discussion forum and so it is highly desirable to detect previous conversation focuses in order to automatically answer queries (Feng et al., 2006).</Paragraph>
    <Paragraph position="2"> Human conversation focus is a hard NLP (Natural Language Processing) problem in general because people may frequently switch topics in a real conversation. The threaded discussions make the problem manageable because people typically focus on a limited set of issues within a thread of a discussion. Current IR (Information Retrieval) techniques are based on keyword similarity measures and do not consider some features that are important for analyzing threaded discussions. As a result, a typical IR system may return a ranked list of messages based on keyword queries even if, within the context of a discussion, this may not be useful or correct.</Paragraph>
    <Paragraph position="3"> Threaded discussion is a special case of human conversation, where people may express their ideas, elaborate arguments, and answer others' questions; many of these aspects are unexplored by traditional IR techniques. First, messages in threaded discussions are not a flat document set, which is a common assumption for most IR systems. Due to the flexibility and special characteristics involved in human conversations, messages within a thread are not necessarily of equal importance. The real relationships may differ from the analysis based on keyword similarity measures, e.g., if a 2 nd message &amp;quot;corrects&amp;quot; a 1 st one, the 2 nd message is probably more important than the 1 st .</Paragraph>
    <Paragraph position="4"> IR systems may give different results. Second, messages posted by different users may have different degrees of correctness and trustworthiness, which we refer to as poster trustworthiness in this paper. For instance, a domain expert is likely to be more reliable than a layman on the domain topic.  In this paper we present a novel feature-enriched approach that learns to detect conversation focus of threaded discussions by combining NLP analysis and IR techniques. Using the graph-based algorithm HITS (Hyperlink Induced Topic Search, Kleinberg, 1999), we conduct discussion analysis taking into account different features, such as lexical similarity, poster trustworthiness, and speech act relations in human conversations. We generate a weighted threaded discussion graph by applying feature-oriented link generation functions. All the features are quantified and integrated as part of the weight of graph edges. In this way, both quantitative features and qualitative features are combined to analyze human conversations, specifically in the format of online discussions.</Paragraph>
    <Paragraph position="5"> To date, it is the first quantitative study to analyze human conversation that focuses on threaded discussions by taking into account heterogeneous evidence from different sources. The study described here addresses the problem of conversation focus, especially for extracting the best answer to a particular question, in the context of an online discussion board used by students in an undergraduate computer science course. Different features are studied and compared when applying our approach to discussion analysis. Experimental results show that performance improvements are significant compared with the baseline system.</Paragraph>
    <Paragraph position="6"> The remainder of this paper is organized as follows: We discuss related work in Section 2. Section 3 presents thread representation and the weighted HITS algorithm. Section 4 details feature-oriented link generation functions. Comparative experimental results and analysis are given in Section 5. We discuss future work in Section 6.</Paragraph>
  </Section>
class="xml-element"></Paper>