<?xml version="1.0" standalone="yes"?>
<Paper uid="P05-1017">
  <Title>Extracting Semantic Orientations of Words using Spin Model</Title>
  <Section position="3" start_page="0" end_page="133" type="relat">
    <SectionTitle>
2 Related Work
</SectionTitle>
    <Paragraph position="0"> Turney and Littman (2003) proposed two algorithms for extraction of semantic orientations of words. To calculate the association strength of a word with positive (negative) seed words, they used the number of hits returned by a search engine, with a query consisting of the word and one of seed words (e.g., &amp;quot;word NEAR good&amp;quot;, &amp;quot;word NEAR bad&amp;quot;). They regarded the difference of two association strengths as a measure of semantic orientation. They also proposed to use Latent Semantic Analysis to compute the association strength with seed words. An empirical evaluation was conducted on 3596 words extracted from General Inquirer (Stone et al., 1966).</Paragraph>
    <Paragraph position="1"> Hatzivassiloglou and McKeown (1997) focused on conjunctive expressions such as &amp;quot;simple and  well-received&amp;quot; and &amp;quot;simplistic but well-received&amp;quot;, where the former pair of words tend to have the same semantic orientation, and the latter tend to have the opposite orientation. They first classify each conjunctive expression into the same-orientation class or the different-orientation class. They then use the classified expressions to cluster words into the positive class and the negative class. The experiments were conducted with the dataset that they created on their own. Evaluation was limited to adjectives.</Paragraph>
    <Paragraph position="2"> Kobayashi et al. (2001) proposed a method for extracting semantic orientations of words with bootstrapping. The semantic orientation of a word is determined on the basis of its gloss, if any of their 52 hand-crafted rules is applicable to the sentence.</Paragraph>
    <Paragraph position="3"> Rules are applied iteratively in the bootstrapping framework. Although Kobayashi et al.'s work provided an accurate investigation on this task and inspired our work, it has drawbacks: low recall and language dependency. They reported that the semantic orientations of only 113 words are extracted with precision 84.1% (the low recall is due partly to their large set of seed words (1187 words)). The hand-crafted rules are only for Japanese.</Paragraph>
    <Paragraph position="4"> Kamps et al. (2004) constructed a network by connecting each pair of synonymous words provided by WordNet (Fellbaum, 1998), and then used the shortest paths to two seed words &amp;quot;good&amp;quot; and &amp;quot;bad&amp;quot; to obtain the semantic orientation of a word. Limitations of their method are that a synonymy dictionary is required, that antonym relations cannot be incorporated into the model. Their evaluation is restricted to adjectives. The method proposed by Hu and Liu (2004) is quite similar to the shortest-path method. Hu and Liu's method iteratively determines the semantic orientations of the words neighboring any of the seed words and enlarges the seed word set in a bootstrapping manner.</Paragraph>
    <Paragraph position="5"> Subjective words are often semantically oriented.</Paragraph>
    <Paragraph position="6"> Wiebe (2000) used a learning method to collect subjective adjectives from corpora. Riloff et al. (2003) focused on the collection of subjective nouns.</Paragraph>
    <Paragraph position="7"> We later compare our method with Turney and Littman's method and Kamps et al.'s method.</Paragraph>
    <Paragraph position="8"> The other pieces of research work mentioned above are related to ours, but their objectives are different from ours.</Paragraph>
  </Section>
class="xml-element"></Paper>