<?xml version="1.0" standalone="yes"?>
<Paper uid="W04-3256">
  <Title>Multi-document Biography Summarization</Title>
  <Section position="3" start_page="0" end_page="0" type="intro">
    <SectionTitle>
2 Recent Developments
</SectionTitle>
    <Paragraph position="0"> Two trends have dominated automatic summarization research (Mani, 2001). One is the work focusing on generating summaries by extraction, which is finding a subset of the document that is indicative of its contents (Kupiec et al., 1995) using &amp;quot;shallow&amp;quot; linguistic analysis and statistics. The other influence is the exploration of Figure 1. Overall design of the biography summarization system.</Paragraph>
    <Paragraph position="1"> &amp;quot;deeper&amp;quot; knowledge-based methods for condensing information. Knight and Marcu (2000) equate summarization with compression at sentence level to achieve grammaticality and information capture, and push a step beyond sentence extraction. Many systems use machine-learning methods to learn from readily aligned corpora of scientific articles and their corresponding abstracts. Zhou and Hovy (2003) show a summarization system trained from automatically obtained text-summary alignments obeying the chronological occurrences of news events.</Paragraph>
    <Paragraph position="2"> MDS poses more challenges in assessing similarities and differences among the set of documents. The simple idea of extract-andconcatenate does not respond to problems arisen from coherence and cohesion. Barzilay et al.</Paragraph>
    <Paragraph position="3"> (1999) introduce a combination of extracted similar phrases and a reformulation through sentence generation. Lin and Hovy (2002) apply a collection of known single-document summarization techniques, cooperating positional and topical information, clustering, etc., and extend them to perform MDS.</Paragraph>
    <Paragraph position="4"> While many have suggested that conventional MDS systems can be applied to biography generation directly, Mani (2001) illustrates that the added functionality of biographical MDS comes at the expense of a substantial increase in system complexity and is somewhat beyond the capabilities of present day MDS systems. The discussion was based in part on the only known MDS biography system (Schiffman et al., 2001) that uses corpus statistics along with linguistic knowledge to select and merge description of people in news. The focus of this work was on synthesizing succinct descriptions of people by merging appositives from semantic processing using WordNet (Miller, 1995).</Paragraph>
  </Section>
class="xml-element"></Paper>