Test suite amelioration using case based reasoning for an agent-based system

Abstract

The paper elaborates on the principle of Test Case reuse by following the Case Based Reasoning Method and puts forward an automatic Test Case generation approach which minimizes the human burden in generating all the possible combinations of the test cases. Software test selects the input data to control the test program and observe the outcome after test execution. The Testing Engineers analyze the result and trace the bug after comparing the expected result and test result. In the software life cycle, testing is an important phase and takes up 30-50% cost. The main theme behind the paper is to reduce the cost and time by at least 10% of the overall cost which is present already. The designing of an agent environment is preferred for the constructing a software because they can work arbitrarily without any intervention, even if it has some obstacles it finds an alternate path for itself without human interference to reach the goal by using its intelligence. There are several methodologies available for the design, implementation and communication of multi agent environment. But the testing of the agents are not been dealt from time immemorial. We developed a Multi-Agent System (MAS) for e-learning application using MASE (Multi-Agent System Engineering) methodology for experimentation. © 2012 Springer-Verlag.