<?xml version="1.0" standalone="yes"?>
<Paper uid="W00-0107">
  <Title>A Measure of Semantic Complexity for Natural Language Systems</Title>
  <Section position="8" start_page="45" end_page="45" type="concl">
    <SectionTitle>
8 Conclusion
</SectionTitle>
    <Paragraph position="0"> This paper describes a way to organize the objects, attributes, classes, and relationships in a domain and to use these classifications to define a semantic domain complexity. This measurement, along with a syntactic complexity measurement, will give natural language programmers a way to quantify the complexity of a given domain in terms of real-world costs: cost of software, reliability, accuracy, and execution time. After defining a syntactic complexity measure, domains can be analyzed against these real costs to be sure that the measure is accurate. Such a measure will allow natural language systems programmers a way to analyze domains and estimate the costs of building a natural language system beforehand, based on the domain's semantic and syntactic constraints. A standard complexity measure will also allow a comparison of different language processors' ability to handle more and more complex domains and quantify the abilities of the current state of the art in natural language processors.</Paragraph>
  </Section>
class="xml-element"></Paper>