Abstract | ||
---|---|---|
Test oracles are usually used to evaluate the behaviors of systems under test to reveal faults. In a typical conformance testing scenario, a test oracle is a deterministic finite state machine (FSM). However, uncertainty occurring in the design of an oracle may result in a set of potential candidate oracles which can compactly be represented by a nondeterministic FSM thus modelling an imprecise test oracle. In the context of testing deterministic systems, such an oracle should ideally be reduced to a precise, i.e., deterministic oracle. We elaborate two scenarios for dealing with imprecise test oracles that involve a domain expert playing the role of an "ultimate" oracle. In the first scenario, the expert chooses a right oracle by inspecting the generated tests differentiating all potential precise oracles which can be derived from a given imprecise one. In the second scenario, the expert evaluates tests one by one while they are generated, and the imprecise oracle is iteratively reduced until a single precise oracle remains, if at all. |
Year | DOI | Venue |
---|---|---|
2019 | 10.1109/ICSTW.2019.00029 | 2019 IEEE International Conference on Software Testing, Verification and Validation Workshops (ICSTW) |
Keywords | Field | DocType |
Uncertainty,Conferences,Test pattern generators,Natural languages,Maintenance engineering | Nondeterministic algorithm,Subject-matter expert,Deterministic finite automaton,Computer science,Oracle,Theoretical computer science,Test pattern generators,Conformance testing,Natural language,Maintenance engineering | Conference |
ISSN | ISBN | Citations |
2159-4848 | 978-1-7281-0888-9 | 0 |
PageRank | References | Authors |
0.34 | 0 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Omer Nguena-Timo | 1 | 19 | 4.77 |
A. Petrenko | 2 | 565 | 31.37 |
Ramesh, S. | 3 | 144 | 19.02 |