Surfacing Ambiguity in Natural Language Requirements.
Published on 2. August 2001
Book
Paperback/Softback
342 pages
978-3-8167-5903-4 (ISBN)
Description
Anforderungsdokumente werden häufig in natürlicher Sprache formuliert, da spezialisierte Sprachen nicht von allen Beteiligten beherrscht werden. Allerdings ist natürliche Sprache inhärent mehrdeutig. Diese Dissertation adressiert das Problem der Aufdeckung von Mehrdeutigkeiten in natürlichsprachlichen Anforderungen, um spätere Softwarefehlverhalten aufgrund von Fehlinterpretationen von Anforderungen zu vermeiden. Mehrere empirisch validierte Techniken zur Aufdeckung von Mehrdeutigkeiten werden für Anforderungsanalyseprozesse unterschiedlichen Reifegrads vorgestellt. Eine Vorgehensweise für die Entwicklung weiterer Techniken wird vorgeschlagen. Weiterhin gibt die Dissertation eine umfassende Definition der verschiedenen Facetten von Mehrdeutigkeit, da gegenwärtige Definitionen in der Literatur widersprüchlich und unvollständig sind.
Industrial requirements documents are written usually in natural language, which is easy to use but also inherently ambiguous. This PhD thesis addresses the problem of detecting ambiguities in natural language requirements in order to avoid software failures due to misinterpretations. A set of empirically validated techniques for detecting ambiguities is presented, which allow improving requirements engineering processes of different maturity. An approach is suggested for developing additional techniques if the provided ones do not fit the situation at hand. New insights are provided on the problems that defects cause when formal specifications are developed from an informal statement of requirements. Also, the thesis provides a comprehensive understanding of the various facets of ambiguity. This understanding clarifies the currently ambiguous use of the term in the literature.
Industrial requirements documents are written usually in natural language, which is easy to use but also inherently ambiguous. This PhD thesis addresses the problem of detecting ambiguities in natural language requirements in order to avoid software failures due to misinterpretations. A set of empirically validated techniques for detecting ambiguities is presented, which allow improving requirements engineering processes of different maturity. An approach is suggested for developing additional techniques if the provided ones do not fit the situation at hand. New insights are provided on the problems that defects cause when formal specifications are developed from an informal statement of requirements. Also, the thesis provides a comprehensive understanding of the various facets of ambiguity. This understanding clarifies the currently ambiguous use of the term in the literature.