Strategy and Domains Adopted In Data Analysis

Exploring strategy and domain adoption in data analysis

Abstract

Less rigorous strategies are practiced widely in testing.Here we refer to such old favorites as boundary testing, testing zero, one, andmany occurrences of some particular phenomenon, and other standard practicesgiven some knowledge of the system specifics, data types, and operators. Theseadapt to the specification level very easily the only transition required isworking with the notation of the specification rather than that of theimplementation as is usually done. Conclusion Our experiments using testing strategies at thespecification level led us to develop two new specification-based testingstrategies. The first, domain propagation, is an extension of partitiontesting. The second, specification mutation, is an adaptation of the existingimplementation-based mutation testing technique.