MACHINE LEARNING BASED METHODOLOGY FOR TESTING OBJECT ORIENTED APPLICATIONS

N Kannadhasan, B Uma Maheswari
2015 unpublished
The objective of software testing is to find the maximum number of errors as early as possible in the software development phase. Software testing ensures the quality of the source code and testing authenticates the incorrectness of the program. The detection of flaws in the C++ source code which leads to logical errors are identified. The code is improved by automatically detecting the defects which are not detected by the compiler. The users are relieved from the liability of detecting these
more » ... of detecting these defects. These frameworks of defects are automatically extracted from the C++ source code by analyzing the object oriented applications. The piece of code contains defect are found and feature values are assigned for training the system. The feature values of defects and non-defect programs are converted into attribute relation format file which constitutes the data set. Machine learning algorithm was used to classify the fault and non-fault statements in the object oriented applications.
fatcat:4zkfhdgenfedbiuvt5ufihb4we