<?xml version="1.0" standalone="yes"?>
<Paper uid="P05-1050">
  <Title>Domain Kernels for Word Sense Disambiguation</Title>
  <Section position="8" start_page="407" end_page="409" type="concl">
    <SectionTitle>
5 Conclusion and Future Works
</SectionTitle>
    <Paragraph position="0"> In this paper we presented a supervised algorithm for WSD, based on a combination of kernel functions. In particular we modeled domain and syntagmatic aspects of sense distinctions by de ning respectively domain and syntagmatic kernels. The Domain kernel exploits Domain Models, acquired from external untagged corpora, to estimate the similarity among the contexts of the words to be disambiguated. The syntagmatic kernels evaluate the similarity between collocations.</Paragraph>
    <Paragraph position="1"> We evaluated our algorithm on several Senseval- null the inter annotator agreement, the F1 of the best system at Senseval-3, the F1 of Kwsd, the F1 of Kprimewsd, DM+ (the improvement due to DM, i.e. Kprimewsd Kwsd).</Paragraph>
    <Paragraph position="2">  of our system outperforms the inter annotator agreement in both English and Spanish, achieving the upper bound performance.</Paragraph>
    <Paragraph position="3"> We demonstrated that using external knowledge  ple task.</Paragraph>
    <Paragraph position="4"> inside a supervised framework is a viable methodology to reduce the amount of training data required for learning. In our approach the external knowledge is represented by means of Domain Models automat- null ically acquired from corpora in a totally unsupervised way. Experimental results show that the use of Domain Models allows us to reduce the amount of training data, opening an interesting research direction for all those NLP tasks for which the Knowledge Acquisition Bottleneck is a crucial problem. In particular we plan to apply the same methodology to Text Categorization, by exploiting the Domain Kernel to estimate the similarity among texts. In this implementation, our WSD system does not exploit syntactic information produced by a parser. For the future we plan to integrate such information by adding a tree kernel (i.e. a kernel function that evaluates the similarity among parse trees) to the kernel combination schema presented in this paper. Last but not least, we are going to apply our approach to develop supervised systems for all-words tasks, where the quantity of data available to train each word expert classi er is very low.</Paragraph>
  </Section>
class="xml-element"></Paper>