Comparative Study of Software Defect Prediction and Analysis the Class using Machine Learning Method
V. Ruckmani1, S. Prakasam2

1V. Ruckmani*, Assistant Professor, Voorhees College Vellore, Anna Salai, Kosapet, Vellore, Tamil Nadu, India.
2S. Prakasam, Associate Professor, Scsvmv University, Enathur, Kanchipuram, Tamil Nadu, India.

Manuscript received on June 08, 2020. | Revised Manuscript received on June 25, 2020. | Manuscript published on June 30, 2020. | PP: 1313-1318 | Volume-9 Issue-5, June 2020. | Retrieval Number: E1161069520/2020©BEIESP | DOI: 10.35940/ijeat.E9957.069520
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (

Abstract: An automatic mode that increases sample stability is checked to verify the software design. Predict software flaws are the main focus of the engineering department. Computational software engineering is one of the active study areas of a software flaw. Depending on the metric, software quality and the efficient allocation of volume resources can easily improve defect quality, thus reducing costs. Many data mining and datasets can be used to store defect prediction software. Machine learning software defect prediction technology is an important branch of the computer. Therefore, in this method is to develop the defect prediction obtained by the design of selected class function metrics to create an effective error finding model. Various models have been proposed to reflect the changing changes in the software product’s defect prediction index. These models also validate the data of the corresponding software module. The software defect analysis uses various software products for performance metrics to predict. It helps to find a different relationship between software volume and error size. Object classes are the user interface components in interactive applications. The control of the function property value assigned to the parsing code. The machine learning logic to detect errors due to defects. Advanced defect prediction models use different methods of performance class and function to evaluate. It provides a valid defect prediction for the defect identification code. This information is implemented in application software to improve predictive error classes and merit function code. 
Keywords: Computational software engineering, identification, application software