<?xml version="1.0" standalone="yes"?>
<Paper uid="H91-1063">
  <Title>SESSION 11 - NATURAL LANGUAGE III</Title>
  <Section position="1" start_page="0" end_page="0" type="intro">
    <SectionTitle>
INTRODUCTION
</SectionTitle>
    <Paragraph position="0"> The five papers in this session, as well as the ten papers in the other two natural language sessions, can be classified into three broad categories: (1) statistical approaches to natural language processing and the automatic acquisition of linguistic structure (2 out of 5 papers in this session; 8 out of 15 overall); (2) robust processing of texts by combining multiple partial analyses (2 out of 5; 5 out of 15); and (3) fundamental issues in linguistic analysis which will become bottlenecks to processing of more complex texts and interactive dialogues (1 out of 5; 2 out of 15). It is quite remarkable that the largest two of these categories, (1) and (2), would probably not been represented in an NLP session at such a meeting just three or four years ago. From the papers sampled here, it is clear that a revolution in natural language processing is occurring and that the strong surge of work in these two areas is closely linked: both of the text analysis systems discussed in this session bring together statistical and non-statistical techniques. Both the rapid change in research direction and the results achieved are surprising and worthy of special note.</Paragraph>
    <Paragraph position="1"> The overview that follows draws out some more detailed cornrnonalities of the papers in this session, and then presents a brief tutorial introduction to two of the problems which several of these papers address.</Paragraph>
  </Section>
class="xml-element"></Paper>