<?xml version="1.0" standalone="yes"?>
<Paper uid="N06-2004">
  <Title>Measuring Semantic Relatedness Using People and WordNet</Title>
  <Section position="3" start_page="0" end_page="13" type="intro">
    <SectionTitle>
2 Data
</SectionTitle>
    <Paragraph position="0"> Aiming at reader-based exploration of lexical cohesion in texts, Beigman Klebanov and Shamir conducted an experiment with 22 students, each reading 10 texts: 3 news stories, 4 journalistic and 3 fiction pieces (Beigman Klebanov and Shamir, 2006). People were instructed to read the text first, and then go over a separately attached list of words in order of their appearance in the text, and ask themselves, for every newly mentioned concept, &amp;quot;which previously mentioned concepts help the easy accommodation of the current concept into the evolving story, if indeed it is easily accommodated, based on the common knowledge as perceived by the annotator&amp;quot; (Beigman Klebanov and Shamir, 2005); this preceding helper concept is called an anchor. People were asked to mark all anchoring relations they could find.</Paragraph>
    <Paragraph position="1"> The rendering of relatedness between two concepts is not tied to any specific lexical relation, but rather to common-sense knowledge, which has to do with &amp;quot;knowledge of kinds, of associations, of typical situations, and even typical utterances&amp;quot;.2 The phenomenon is thus clearly construed as much broader than degree-of-synonymy.</Paragraph>
    <Paragraph position="2"> Beigman Klebanov and Shamir (2006) provide reliability estimation of the experimental data using ness&amp;quot; (Hirst and Budanitsky, 2005); &amp;quot;To our knowledge, no datasets are available for validating the results of semantic relatedness metric&amp;quot; (Gurevych, 2005).</Paragraph>
    <Paragraph position="3">  statistical analysis and a validation experiment, identifying reliably anchored items with their strong anchors, and reliably un-anchored items. Such analysis provides high-validity data for classification; however, much of the data regarding intermediate degrees of relatedness is left out.</Paragraph>
  </Section>
class="xml-element"></Paper>