IEEE Project Abstract

Software maintainability predicts changes or failures that may occur in software after it hasbeen deployed. Since it deals with the degree to which an application may be understood, repaired orenhanced, it also takes into account the overall cost of the project. In the past several measures have beentaken into account for predicting metrics that influence software maintainability. However, deep learning isyet to be explored for the same. In this paper, we perform Deep Learning for Software MaintainabilityMetrics Prediction on a large number of datasets. Unlike the previous research works, we have relied onlarge datasets from 299 software’s and subsequently applied various metrics and functions to the same.Twenty-nine object-oriented metrics have been considered along with their impact on softwaremaintainability of open source software. Five Machine Learning algorithms namely Ridge Regression withVariable Selection, Decision Tree, Quantile Regression Forest, Support Vector Machine and PrincipalComponent Analysis have been applied to the original datasets, as well as to the refined datasets. It wasfound that our study provides results in form of metrics that may be used in the prediction of softwaremaintenance and the proposed Deep Learning model outperforms all of the other methods that wereconsidered. Further, the results of experiment affirm the efficiency of the proposed Deep Learning modelfor software maintainability prediction.