<?xml version="1.0" standalone="yes"?>
<Paper uid="W03-1306">
  <Title>Boosting Precision and Recall of Dictionary-Based Protein Name Recognition</Title>
  <Section position="6" start_page="0" end_page="0" type="relat">
    <SectionTitle>
6 Related Work
</SectionTitle>
    <Paragraph position="0"> Kazama et al. (2002) reported an F-measure of 56.5% on the GENIA corpus (Version 1.1) using Support Vector Machines. Collier et al. (2001) reported an F-measure of 75.9% evaluated on 100 MEDLINE abstracts using a Hidden Markov Model.</Paragraph>
    <Paragraph position="1"> These research efforts are machine learning based and do not provide ID information of recognized terms.</Paragraph>
    <Paragraph position="2"> Krauthammer et al. (2000) proposed a dictionary-based gene/protein name recognition method. They used BLAST for approximate string matching by mapping sequences of text characters into sequences of nucleotides that can be processed by BLAST.</Paragraph>
    <Paragraph position="3"> They achieved a recall of 78.8% and a precision of 71.1% by a partial match criterion, which is less strict than our exact match criterion.</Paragraph>
  </Section>
class="xml-element"></Paper>