Authors:
Dmitrij Koznov
1
;
Vasily Malinov
2
;
Eugene Sokhransky
2
and
Marina Novikova
2
Affiliations:
1
St.-Petersburg State University, Faculty of Mathmatics and Mechanic, Russian Federation
;
2
DataArt, Inc., United States
Keyword(s):
Knowledge Management, Model-Based Testing, Partial Specifications, Requirement Recovering, Architecture Recovering, UniTesk
Related
Ontology
Subjects/Areas/Topics:
Artificial Intelligence
;
Communication, Collaboration and Information Sharing
;
KM Strategies and Implementations
;
Knowledge Management and Information Sharing
;
Knowledge-Based Systems
;
Symbolic Systems
Abstract:
This paper offers a knowledge management method for industrial model-based testing, which based on partial specifications and “attached” to the software development process that uses it. Partial specification means formal description of considerable/potentially problematic properties of a system, and is used for further automated testing. That allows reducing expenses of testing compared to developing full formal specifications. The “attached” nature of the method means that the team of testers can work independently of the basic process, without imposing on it any specific limitations connected with model-based testing. The method intends for lightweight processes where a lack of documentation and formal described requirements are absent. The paper also presents approbation of the method while testing an industrial Web-application by means of model-based testing technology UniTesk.