AN AUTOMATED GENETIC AND DEEP LEARNING FUTURE DEFECT PREDICTOR FOR REWORK REDUCTION

  • P. Patchaiammal, G. Sundar, Dr. R. Thirumalaiselvi,
Keywords: Machine Learning (ML), Deep Learning (DL), Rework, Software Development, Predictor, Activation Function, Optimizer, GitHub Repository, Python, Tkinter, Keras.

Abstract

An imperfect feature of software function after release is known as defect.  Defects are the error found after the application finished the production, but these defects are mainly created by development stage faults.It makes the software product difficult with external behaviour or internal features. These defects in real-time software are abundant, so the analysis of the defects will be more challenging to the developers. To identify the root cause of the defect, a developer needs to identify the defect in early stage of development. The rework reduction will be achieved only by maintaining classified defect taxonomy set.To form defect taxonomy, first one has to identify whether the defects are predictable or not. This research work will help to develop an automated defect predictor using deep learning technique.

Published
2021-07-23
How to Cite
Dr. R. Thirumalaiselvi, P. P. G. S. (2021). AN AUTOMATED GENETIC AND DEEP LEARNING FUTURE DEFECT PREDICTOR FOR REWORK REDUCTION. Design Engineering, 4487- 4496. Retrieved from http://www.thedesignengineering.com/index.php/DE/article/view/2896
Section
Articles