|
|
 |
| |
|
|
|
Investigation Analysis for Software Fault Prediction using Error Probabilities and Integral Methods |
|
|
|
PP: 205-210 |
|
|
|
doi:10.18576/amis/13S121
|
|
|
|
Author(s) |
|
|
|
S. Karuppusamy,
G. Singaravel,
|
|
|
|
Abstract |
|
|
| In this paper, in-depth analysis of faults in the code phase is detected through integral methods that identify the error in software. The repositories of the data set are collected during the software product development life cycle model, which is then integrated with a machine learning algorithm namely Bayesian decision theory to detect the error probabilities and to predict unbound error during the prediction of the software faults. In prior, the faults are predicted in repository for a given data set using error probability and error integral method that identify the probability of error and correction, which is then applied with Gaussian method to find the levels of the error probability with minimum and maximum integral of acceptable faults in the repository.
|
|
|
|
|
|
|
|
