<?xml version="1.0" standalone="yes"?>
<Paper uid="C00-1044">
  <Title>Effects of Adjective Orientation and Gradability on Sentence Subjectivity</Title>
  <Section position="6" start_page="303" end_page="304" type="evalu">
    <SectionTitle>
5 Conclusion and Future Work
</SectionTitle>
    <Paragraph position="0"> This paper presents an analysis of different adjective features for predicting subjectivity, showing that tlmy are more precise than those previously used for this task. Wc establish that lexical semantic features such as semantic orientation and gradability determine in large part the subjectivity status of sentences in which they appear. We also present an automatic meflmd for extracting gradability values reliably, complementing earlier work on semantic orientation and dynamic adjectives.</Paragraph>
    <Paragraph position="1"> In addition to finding more precise features for automarie subjectivity recognition, this kind of analysis could help efforts to encode subjective features in ontologies such as those described in (Knight and Luk, 1994; Mahesh and Nirenburg, 1995; Hovy, 1998). These ontologies are useful for many NLP tasks, such as machine translation, word-sense disambiguation, and generation. Some subjective features are included in existing ontologies (for example, Mikrokosmos (Mahesh and  Nirenburg, 1995) includes atlitude slots). Our corpus-based methods could help in idenlifying more or exlending their coverage.</Paragraph>
    <Paragraph position="2"> To be able to use automatic subjectivily recognition in texl-processing applications, good ch,cs o1' sub.icclivity mttst be found. The features developed in lhis paper are not only good clues of subjectivity, lhey can be Mentilied automatically from corpora (see (Hatzivassiloglou and McKeown, 1997), and Section 3 in the present paper). In fact, the results in &amp;quot;Iable 3 show that the predictability of the automatically determined gradability and polarity sets is better than or at least comparable to the predictability of the manually determined sets. Thus, tile oriented and gradable adjectives in the particular application genre can be idenlified fo,&amp;quot; use in subjectivity recognition.</Paragraph>
    <Paragraph position="3"> Ou, efforts in this paper are largely exploratory, aiming to establish correlations among tim wlrious features examined. In related work, we have begun to incorporale the features developed herc into systems for recognizing flames and mining reviews in lnternel forums, extending subjectivity judgments froth the sentence to the document level. In addition, we are seeking ways lo extend the orientation and gradability methods so that individual word occurrences, rather than word lypes, are characterized as oriented or gradable. We also pla n l{7 incorporate the new features presented here in machine learning models for tile prediction of subjectivity (e.g., (Wiebe ct al., 1999)) and lest lheir interaclions wilh olhcr proposed features.</Paragraph>
  </Section>
class="xml-element"></Paper>