<?xml version="1.0" standalone="yes"?>
<Paper uid="W06-1908">
  <Title>Dialogue based Question Answering System in Telugu</Title>
  <Section position="2" start_page="0" end_page="53" type="intro">
    <SectionTitle>
1 Introduction
</SectionTitle>
    <Paragraph position="0"> Ever since Question Answering (QA) emerged as an active research field, the community has slowly diversified question types, increased question complexity, and refined evaluation metrics, as reflected by the TREC (Text Retrieval Conference) QA track (Voorhees, 2004). Several QA systems have responded to these changes in the nature of the QA task by incorporating various knowledge resources (Hovy et al., 2002), handling of additional types of questions tapping into external data sources such as web, encyclopedia, and databases in order to find the answer candidates, which may then be located in the specific corpus being searched (Xu et al., 2003).</Paragraph>
    <Paragraph position="1"> The most popular classes of technique for QA are open-domain and restricted-domain (Diekema et al., 2004, Doan-Nguyen et al., 2004). These two domains use thesauri and lexicons in classifying documents and categorizing the questions. Open domain question answering deals with questions about nearly everything and can only rely on general ontology. It has become a very active research area over the past few years. On the other hand, Restricted-domain question answering (RDQA) deals with questions under a specific domain. If we create such a RDQA interface for structured e.g. relational database, we call it as Natural language interface to database system (NLIDB) (Androutsopoulos et al., 1995), where it allows the user to access the information stored in database by typing requests expressed in some natural language.</Paragraph>
    <Paragraph position="2"> RDQA has a long history, beginning with systems working over databases (e.g., BASEBALL (Green et al., 1961), and LUNAR (woods et al., 1972)).</Paragraph>
    <Paragraph position="3"> In practice, current QAs can only understand limited subsets of natural language. Therefore, some training is still needed to teach the end-user what kinds of questions the system can or cannot understand. There are kinds of questions (e.g.</Paragraph>
    <Paragraph position="4"> questions involving negation, or quantification) that can be easily expressed in natural language, but that seem difficult (or at least tedious) to express using graphical or form based interfaces.</Paragraph>
    <Paragraph position="5"> Anaphoric and elliptical expressions are also handled by the QA systems. In recent years a large part of the research in QAs has been devoted to portability, i.e., to the design of QAs that can be used in different knowledge domains</Paragraph>
    <Section position="1" start_page="0" end_page="53" type="sub_section">
      <SectionTitle>
Rami Reddy Nandi Reddy
</SectionTitle>
      <Paragraph position="0"> Dept. of Comp. Sc. &amp; Engg,  (Knowledge domain portability), with different underlying Database Management System (DBMS) (DBMS portability), or even with different natural languages (Natural language portability). There is a growing body of research on integrating speech recognition, robust interpretation with the goal being to implement systems that engage users in spoken dialogue to help them perform certain tasks. We expect that this line of research will have a significant influence on future QAs, giving rise to systems that will allow users to access databases by spoken dialogue, in situations for which graphic and form-based interfaces are difficult to use.</Paragraph>
      <Paragraph position="1"> A practical question answering system in restricted domain (Hoojung et al., 2004) and our system handles user questions similarly. However, our system extracts the information from a relational database. Moreover, our system keeps track of user dialogue and handles clarifications, elaborations and confirmations needed from the user with respect to the query. Along with it returns natural language answer in user-friendly format.</Paragraph>
    </Section>
  </Section>
class="xml-element"></Paper>