<?xml version="1.0" standalone="yes"?>
<Paper uid="W05-0205">
  <Title>Towards Intelligent Search Assistance for Inquiry-Based Learning</Title>
  <Section position="3" start_page="0" end_page="25" type="relat">
    <SectionTitle>
2 Related Work
</SectionTitle>
    <Paragraph position="0"> In Information Retrieval field, many algorithms based on relevance feedback are proposed (Buckley, et al., 1994; Salton and Buckley, 1990). However, current general web search engines are still unable to interactively improve research results. In NLP domain, there are considerable efforts on Question Answering systems that attempt to answer a question by returning concise facts.</Paragraph>
    <Paragraph position="1"> While some QA systems are promising (Harabagiu, et al., 2000; Ravichandran and Hovy, 2002), they can only handle factual questions as in TREC (Voorhees, 2001), and the context for the whole task is largely not considered. There are proposals on using context in search. Huang et al (2001) proposed a term suggestion method for interactive web search. More existing systems that utilize contextual information in search are reviewed by Lawrence (2000). However, one problem is that &amp;quot;context&amp;quot; is defined differently in each  study. Few attempts target at inquiry-based learning, which has some unique features, e.g., DQ/SQ. We are developing an OnLine Inquiry Search Assistance (OLISA). OLISA applies Natural Language Processing (NLP) and Information Retrieval (IR) techniques to provide students query term suggestions and re-rank results returned from search engines by the relevance to the current query as well as to the DQ. OLISA is not a built-in component of IdeaKeeper, but can be very easily plugged into IdeaKeeper or other OIBL systems as a value-added search agent. The main advantage of OLISA is that it utilizes the context of the whole learning task. Our pilot study demonstrated that it is a simple and effective initiative toward automatically improving the quality of web search in OIBLE.</Paragraph>
  </Section>
class="xml-element"></Paper>