<?xml version="1.0" standalone="yes"?>
<Paper uid="C00-1044">
  <Title>Effects of Adjective Orientation and Gradability on Sentence Subjectivity</Title>
  <Section position="2" start_page="0" end_page="299" type="intro">
    <SectionTitle>
1 Introduction
</SectionTitle>
    <Paragraph position="0"> In recent years, computalional tcchniqt,es for the determination of &amp;:deal semantic features have been proposed and ewdualed. Such features include sense, register, domain spccilicity, pragmatic restrictions on usage, scnlanlic markcdncss, and orientation, as well as automatically ictcnlifiecl links between words (e.g., semantic rclalcdhess, syllollynly, antonylny, and tneronymy). Aulomalically learning features of this type from hugc corpora allows the construction or augmentation of lexicons, and the assignment of scmanlic htbcls lo words and phrases in running text. This information in turn can bc used to help dcterlninc addilional features at the Itxteal, clause, sentence, or document level.</Paragraph>
    <Paragraph position="1"> Tiffs paper explores lira benelits that some lexical features of adjectives offer l'or the prediction of a contexlual sentence-level feature, suOjectivity. Subjectivity in natural language re\['crs to aspects of language used to express opinions and ewfluations. The computatiomtl task addressed here is to distinguish sentences used to present opinions and other tbrms of subjectivity (suOjective sentences, e.g., &amp;quot;At several different layers, it's a fascinating title&amp;quot;) from sentences used to objectively present factual information (objective sentences, e.g., &amp;quot;Bell industries Inc. increased its quarterly to 10 cents from 7 cents a share&amp;quot;).</Paragraph>
    <Paragraph position="2"> Much research in discourse processing has focused on task-oriented and insmmtional dialogs. The task addressed here comes to the fore in other genres, especially news reporting and lnternet lorums, in which opinions of various agents are expressed and where subjectivity judgements couht help in recognizing inllammatory ruessages (&amp;quot;llanles') and mining online sources for product reviews. ()thor (asks for whicll subjectivity recognition is potentially very useful include infornmtion extraction and information retrieval. Assigning sub.icctivity labels to documents or portions of documents is an example of non-topical characterixation of information. Current information extraction and rolricval lechnology focuses almost exclusively on lhe subject matter of the documcnls.</Paragraph>
    <Paragraph position="3"> Yet, additiomtl components of a document inllucncc its relevance to imrlicuhu * users or tasks, including, for exalnple, the evidential slatus el: lhc material presented, and attitudes adopted in fawn&amp;quot; or against a lmrticular person, event, or posilion (e.g., articles on a presidenlial campaign wrillen to promote a specific candidate). In summarization, subjectivity judgmcnls could be included in documcllt proiilcs to augment aulomatically produced docunacnt summaries, and to hel l) the user make relevance judgments when using a search engine.</Paragraph>
    <Paragraph position="4"> ()thor work on sub.iectivity (Wicbc et al., 1999; Bruce and Wicbc, 2000) has established a positive and statistically signilicant correlation with the presence of adieclives. PSincc the mere presence of one or iDoi'c adjectives is useful for prcdicling (hat a scntcrtce is subjective we investigate ill this paper (lie cfl'ccts of additional lcxical scmanlic lcalurcs of adjectives that can be automatically learned from corpora. We consider two such l%atures: semantic orientation, which represents an ewdualivc characterization of a word's deviation from the norm for its semantic group (e.g., beauti/'ul is positively oriented, as opposed to ugly); and gradability, which characterizes a word's ability to express a property in wlrying degrees.</Paragraph>
    <Paragraph position="5"> In lira remainder of this paper, we \[irst address adjective orientation in Section 2, summarizing a previously published method for automatically separating oriented adjectives into positive and negative classes. Then, Section 3 presents a novel method for learning gradablc adjectives using a largo corpus and a statistical feature combination naodel. In Section 4, we review earlier experiments on testing subjectivity using wu'ious fcatt, res as predictors, and then present comparative analyses of the effects that orientation and gradability have on our ability to In'edict sentence subjectivity from adjectives. Wc show that both give us higher-quality features for recognizing st@icctive sentences, and conclude by discussing future extensions to Ibis work.</Paragraph>
  </Section>
class="xml-element"></Paper>