An optimized test case generation technique for enhancing state-sensitivity partitioning
Software testing is a vital phase in software development life cycle (SDLC) and its principal element is test case. Test case generation remains the most dominant the research area in software testing. One of the techniques that were proposed for generating test cases is State Sensitivity Partiti...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
2017
|
| Subjects: | |
| Online Access: | http://psasir.upm.edu.my/id/eprint/68748/1/FSKTM%202018%209%20IR.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Software testing is a vital phase in software development life cycle (SDLC) and its
principal element is test case. Test case generation remains the most dominant
the research area in software testing. One of the techniques that were proposed
for generating test cases is State Sensitivity Partitioning (SSP). It aims to avoid
the exhaustive testing of module’s entire states. It partitions the entire data states
based on their sensitivities towards events, conditions and actions. The test data
for SSP is in the form of event sequences. As there is no limit on the number of
events that any sequence can hold, lengthy test cases might be generated.
Besides, no constrains were applied in order to avoid retesting a component that
was already tested. Subsequently, a state explosion might be encountered.
The aim of this study was to address the problem of redundant states encountered
within SSP test cases. An optimization technique was proposed, enSSP, featuring
the generation of optimized test cases. The scope of this work is testing a module
with memory where each module may consist of several programs. The essence
of enSSP is to combine the features of Genetic Algorithm (GA) with a suite
reduction technique to achieve optimization. GA removes redundant states from
test cases while the reduction technique removes redundant sequences from the
suite. Afterwards, a prioritization algorithm used for sorting the test cases so the
first test case detects the highest number of mutants followed by the cases that
kill its live mutants. Experiments were conducted using mutation analysis to
compare the fault detection capabilities of enSSP and SSP. The main interest of
the experiment is to demonstrate the capability of enSSP. With respect to both
quality attributes, the effectiveness and the efficiency, the results indicate that
enSSP is more effective and efficient than SSP. |
|---|