System Error Estimate using Combination of Classification and Optimization Technique

The representation of software must produce flawless significances without any inadequacies. Software imperfection evaluations scheme determines defective mechanisms in software. The eventual creation would have minor or negligible shortcomings to harvest great eminence software. Software quality metrics are a division of software metrics that spotlight the quality aspects of the product. The software flaw prediction system helps in the early discovery of flaws and contributes to talented removal and producing a quality software system through numerous metrics. The aim of the paper was to show how static model of data mining is used to extract defects and the PSO algorithm. Another aim of the research was to develop an optimized software flaw prophecy system on data mining techniques namely Association Rule mining, Decision Tree, Naive Bayes and Classification integrated with Particle Swarm Optimization technique. The proposed software flaw prediction system is deliberated through Data Mining techniques with Particle Swarm Optimization algorithm has been verified and compared the results. This proposed system is very useful to identify the relationships between the quality metrics and the potential defective modules. The optimized data mining systems have pragmatic perfect prediction of these defective modules. In the future, optimized data mining systems can be improved by the use of different platforms and particularly by improving data mining using PSO algorithms. It is necessary to develop algorithms that can identify faults in advance, which will minimize costs and promote the quality of developed software systems. Future optimized data mining systems will improve the relationship between quality metrics and the potential defective modules, which will lead to improved performances, productivity and lower operation costs.