<?xml version="1.0" standalone="yes"?>
<Paper uid="W03-1113">
  <Title>Dynamic Programming Matching for Large Scale Information Retrieval</Title>
  <Section position="6" start_page="0" end_page="0" type="relat">
    <SectionTitle>
5 Related Work
</SectionTitle>
    <Paragraph position="0"> Our proposed technique is a type of DP matching.</Paragraph>
    <Paragraph position="1"> The most typical application of DP matching is gene information research, because DP is effective for gene information matching. However, this system has a very slow processing speed.</Paragraph>
    <Paragraph position="2"> In recent years, advances in this field of research have meant that high-speed systems have been required for gene information retrieval. A high-speed gene information retrieval system called BLAST was developed (Setubal and Meidanis, 2001). BLAST has achieved higher processing speed by using heuristics that specify characteristic gene arrangements, rather than using DP matching.</Paragraph>
    <Paragraph position="3"> In contrast, we have managed to achieve fast match-ing using the DP technique.</Paragraph>
    <Paragraph position="4"> Moreover, in music information retrieval, an error in copying a tune corresponds to a deficit (deletion) and insertion of data. For this reason, a music search engine has been built based on the DP technique (Hu and Dannenberg, 2002). Since there is a great deal of music information available these days, scalability is also an important problem for music information retrieval systems. Our proposed DP method is scalable and can cope with deficits. It therefore has potential applications in music information retrieval.</Paragraph>
  </Section>
class="xml-element"></Paper>