<?xml version="1.0" standalone="yes"?>
<Paper uid="W03-1016">
  <Title>Statistical Acquisition of Content Selection Rules for Natural Language Generation</Title>
  <Section position="2" start_page="0" end_page="0" type="intro">
    <SectionTitle>
1 Introduction
CONTENT SELECTION is the task of choosing the
</SectionTitle>
    <Paragraph position="0"> right information to communicate in the output of a Natural Language Generation (NLG) system, given semantic input and a communicative goal. In general, Content Selection is a highly domain dependent task; new rules must be developed for each new domain, and typically this is done manually. Morevoer, it has been argued (Sripada et al., 2001) that Content Selection is the most important task from a user's standpoint (i.e., users may tolerate errors in wording, as long as the information being sought is present in the text).</Paragraph>
    <Paragraph position="1"> Designing content selection rules manually is a tedious task. A realistic knowledge base contains a large amount of information that could potentially be included in a text and a designer must examine a sizable number of texts, produced in different situations, to determine the specific constraints for the selection of each piece of information.</Paragraph>
    <Paragraph position="2"> Our goal is to develop a system that can automatically acquire constraints for the content selection task. Our algorithm uses the information we learned from a corpus of desired outputs for the system (i.e., human-produced text) aligned against related semantic data (i.e., the type of data the system will use as input). It produces constraints on every piece of the input where constraints dictate if it should appear in the output at all and if so, under what conditions. This process provides a filter on the information to be included in a text, identifying all information that is potentially relevant (previously termed global focus (McKeown, 1985) or viewpoints (Acker and Porter, 1994)). The resulting information can be later either further filtered, ordered and augmented by later stages in the generation pipeline (e.g., see the spreading activation algorithm used in ILEX (Cox et al., 1999)).</Paragraph>
    <Paragraph position="3"> We focus on descriptive texts which realize a single, purely informative, communicative goal, as opposed to cases where more knowledge about speaker intentions are needed. In particular, we present experiments on biographical descriptions, where the planned system will generate short paragraph length texts summarizing important facts about famous people. The kind of text that we aim to generate is shown in Figure 1. The rules that we aim to acquire will specify the kind of information that is typically included in any biography. In some cases, whether Actor, born Thomas Connery on August 25, 1930, in Fountainbridge, Edinburgh, Scotland, the son of a truck driver and charwoman. He has a brother, Neil, born in 1938. Connery dropped out of school at age fifteen to join the British Navy. Connery is best known for his portrayal of the suave, sophisticated British spy, James Bond, in the 1960s. . . .</Paragraph>
    <Paragraph position="4">  the information is included or not may be conditioned on the particular values of known facts (e.g., the occupation of the person being described --we may need different content selection rules for artists than politicians). To proceed with the experiments described here, we acquired a set of semantic information and related biographies from the Internet and used this corpus to learn Content Selection rules.</Paragraph>
    <Paragraph position="5"> Our main contribution is to analyze how variations in the data influence changes in the text. We perform such analysis by splitting the semantic input into clusters and then comparing the language models of the associated clusters induced in the text side (given the alignment between semantics and text in the corpus). By doing so, we gain insights on the relative importance of the different pieces of data and, thus, find out which data to include in the generated text.</Paragraph>
    <Paragraph position="6"> The rest of this paper is divided as follows: in the next section, we present the biographical domain we are working with, together with the corpus we have gathered to perform the described experiments. Section 3 describes our algorithm in detail. The experiments we perform to validate it, together with their results, are discussed in Section 4. Section 5 summarizes related work in the field. Our final remarks, together with proposed future work conclude the paper. null</Paragraph>
  </Section>
class="xml-element"></Paper>