<?xml version="1.0" standalone="yes"?>
<Paper uid="P05-1053">
  <Title>Exploring Various Knowledge in Relation Extraction</Title>
  <Section position="3" start_page="427" end_page="427" type="relat">
    <SectionTitle>
2 Related Work
</SectionTitle>
    <Paragraph position="0"> The relation extraction task was formulated at the  1998) and is starting to be addressed more and more within the natural language processing and machine learning communities.</Paragraph>
    <Paragraph position="1"> Miller et al (2000) augmented syntactic full parse trees with semantic information corresponding to entities and relations, and built generative models for the augmented trees. Zelenko et al (2003) proposed extracting relations by computing kernel functions between parse trees. Culotta et al (2004) extended this work to estimate kernel functions between augmented dependency trees and achieved 63.2 F-measure in relation detection and 45.8 F-measure in relation detection and classification on the 5 ACE relation types. Kambhatla (2004) employed Maximum Entropy models for relation extraction with features derived from word, entity type, mention level, overlap, dependency tree and parse tree. It achieves 52.8 F-measure on the 24 ACE relation subtypes. Zhang (2004) approached relation classification by combining various lexical and syntactic features with bootstrapping on top of Support Vector Machines.</Paragraph>
    <Paragraph position="2"> Tree kernel-based approaches proposed by Zelenko et al (2003) and Culotta et al (2004) are able to explore the implicit feature space without much feature engineering. Yet further research work is still expected to make it effective with complicated relation extraction tasks such as the one defined in ACE. Complicated relation extraction tasks may also impose a big challenge to the modeling approach used by Miller et al (2000) which integrates various tasks such as part-of-speech tagging, named entity recognition, template element extraction and relation extraction, in a single model. This paper will further explore the feature-based approach with a systematic study on the extensive incorporation of diverse lexical, syntactic and semantic information. Compared with Kambhatla (2004), we separately incorporate the base phrase chunking information, which contributes to most of the performance improvement from syntactic aspect. We also show how semantic information like WordNet and Name List can be equipped to further improve the performance. Evaluation on the ACE corpus shows that our system outperforms Kambhatla (2004) by about 3 F-measure on extracting 24 ACE relation subtypes. It also shows that our system outperforms tree kernel-based systems (Culotta et al 2004) by over 20 F-measure on extracting 5 ACE relation types.</Paragraph>
  </Section>
class="xml-element"></Paper>