<?xml version="1.0" standalone="yes"?>
<Paper uid="P05-3017">
  <Title>Supporting Annotation Layers for Natural Language Processing</Title>
  <Section position="3" start_page="0" end_page="0" type="intro">
    <SectionTitle>
1 Introduction
</SectionTitle>
    <Paragraph position="0"> Today most natural language processing (NLP) algorithms make use of the results of previous processing steps. For example, a word sense disambiguation algorithm may combine the output of a tokenizer, a part-of-speech tagger, a phrase boundary recognizer, and a module that classifies noun phrases into semantic categories. Currently there is no standard way to represent and store the results of such processing for efficient retrieval.</Paragraph>
    <Paragraph position="1"> We propose a framework for annotating text with the results of NLP processing and then querying against those annotations in flexible ways. The framework includes a query language and an indexing architecture for efficient retrieval, built on top of a relational database management system (RDBMS). The model allows for both hierarchical and overlapping layers of annotation as well as for querying at multiple levels of description.</Paragraph>
    <Paragraph position="2"> In the remainder of the paper we describe related work, illustrate the annotation model and the query language and describe the indexing architecture and the experimental results, thus showing the feasibility of the approach for a variety of NLP tasks.</Paragraph>
  </Section>
class="xml-element"></Paper>