<?xml version="1.0" standalone="yes"?>
<Paper uid="N03-1004">
  <Title>In Question Answering, Two Heads Are Better Than One</Title>
  <Section position="8" start_page="0" end_page="0" type="concl">
    <SectionTitle>
7 Conclusions
</SectionTitle>
    <Paragraph position="0"> In this paper, we introduced a multi-strategy and multi-source approach to question answering that enables combination of answering agents adopting different strategies and consulting multiple knowledge sources. In particular, we focused on two answering agents, one adopting a knowledge-based approach and one using statistical methods. We discussed our answer resolution component which employs a multi-level combination algorithm that allows for resolution at the question, passage, and answer levels. Best performance using the % correct metric was achieved by the three-level algorithm that combines after each stage, while highest average precision was obtained by a two-level algorithm merging at the question and answer levels, supporting a tightly-coupled design for multi-agent question answering. Our experiments showed that our best performing algorithms achieved a 35.0% relative improvement in the number of correct answers and a 32.8% improvement in average precision on a previously unseen test set.</Paragraph>
  </Section>
class="xml-element"></Paper>