A Probabilistic Parser Applied to Software Testing Documents

Abstract

We describe an approach to training a statistical parser from a bracketed corpus, and demonstrate its use in a software testing application that translates English specifications into an automated testing language. A grammar is not explicitly specified; the rules and contextual probabilities of occurrence are automatically generated from the corpus. The parser is extremely successful at producing and identifying the correct parse, and nearly deterministic in the number of parses that it produces. To compensate for undertraining, the parser also uses general, linguistic subtheories which aid in guessing some types of novel structures. Introduction In constrained domains, natural language processing can often provide leverage. In software testing at AT&T, for example, 20,000 English test cases prescribe the behavior of a telephone switching system. A test case consists of about a dozen sentences describing the goal of the test, the actions to perform, and the conditions to verify. Figu...