Adopting genetic algorithm to enhance state-sensitivity partitioning

• Main Authors: Sultan, Ammar Mohammed, Baharom, Salmi, Abd Ghani, Abdul Azim, Din, Jamilah, Zulzalil, Hazura
• Format: Conference or Workshop Item
• Language: English
• Published: School of Computing, Universiti Utara Malaysia 2015
• Online Access: http://psasir.upm.edu.my/id/eprint/59076/1/PID040.pdf; http://psasir.upm.edu.my/id/eprint/59076/
• Institution: Universiti Putra Malaysia

Software testing requires executing software under test with the intention of finding defects as much as possible. Test case generation remains the most dominant research in software testing. The technique used in generating test cases may lead to effective and efficient software testing process. Many techniques have been proposed to generate test cases. One of them is State Sensitivity Partitioning (SSP) technique. The objective of SSP is to avoid exhaustive testing of the entire data states of a module. In SSP, test cases are represented in the form of sequence of events. Even recognizing the finite limits on the size of the queue, there is an infinite set of these sequences and with no upper bound on the length of such a sequence. Thus, a lengthy test sequence might consist of redundant data states. The existence of the redundant data state will increase the size of test suite and consequently the process of testing will be ineffective. Therefore, there is a need to optimize those test cases generated by the SSP in enhancing its effectiveness in detecting faults. Genetic algorithm (GA) has been identified as the most common potential technique among several optimization techniques. Thus, GA is investigated for the integrating with the existing SSP. This paper addresses the issue on how to represent the states produced by SSP sequences of events in order to be accepted by GA. System ID were used for representing the combination of states variables uniquely and generate the GA initial population.