Coupling and Cohesion Metrics-based Fault Predictions using Machine learning Algorithm

Malware affects the software system and may lead to server failure. It is required to detect the malicious software accurately to avoid the time and cost wastage. Various research works has been introduced earlier for the detection of malicious software. In the existing work, Support Vector Machine (SVM) is introduced for malicious software detection, which cannot perform well where there are error modules in software. It is addressed in this suggested study by developing the Coupling and Cohesion Metrics based Fault Detection (CCMFD). A high - quality service has a low coupling and a high cohesiveness. In this research work, structural measures are mainly examined which comes under group of cohesion measures and comprises deficient of no cohesion in methods and CCBO approaches. Failure situations measures relating to information flow which are used in other techniques. These extracted features will be given as input to the Modified Convolutional Neural Network for software mistake forecasting. The complete analysis of the study is implementing in Java simulator.