Optimisation of software testing using Genetic Algorithm
International Journal of Artificial Intelligence and Soft Computing
Automatic generation of basis test paths using variable length genetic algorithm
Information Processing Letters
| Hi-index | 0.00 |
This paper presents a method for optimizingsoftware testing efficiency by identifying the most errorprone path clusters in a program. We do this bydeveloping variable length Genetic Algorithms thatoptimize and select the software path clusters whichare weighted with sources of error indexes. Althoughvarious methods have been applied to detecting andreducing errors in a whole system, there is littleresearch into partitioning a system into smaller errorprone domains for testing. Exhaustive software testingis rarely possible because it becomes intractable foreven medium sized software. Typically only parts of aprogram can be tested, but these parts are notnecessarily the most error prone. Therefore, we aredeveloping a more selective approach to testing byfocusing on those parts that are most likely to containfaults, so that the most error prone paths can be testedfirst. By identifying the most error prone paths, thetesting efficiency can be increased.