Name
Title | ||
|---|---|---|
Improving Software Maintainability Predictions using Data Oversampling and Hybridized Techniques. |
Abstract | ||
|---|---|---|
Software systems being developed in today’s era are inherently large and complex. Maintaining these software is a big challenge before the software industry. Since software maintenance demands a high cost, this activity becomes even more challenging. In order to reduce the maintenance cost, it becomes crucial to know the maintainability of software modules/ classes in the initial stages of software development. Many efforts have been made to identify the maintainability of modules in the initial development stages in which prediction models have a lot of roles. Prediction models are trained from past historical data and should consist of an adequate number of instances of low maintainability and high maintainability class/modules. But this is not usually the case, and because of this, we are not able to train the prediction models properly. This situation shows data imbalance. In this direction, in this paper, we will handle the imbalanced data problem so that prediction models can be properly trained. We apply four oversampling techniques in this paper and train the prediction models for maintainability by hybridized techniques. The results of the paper advocate the effectiveness of examined oversampling techniques along with the hybridized classification techniques to develop competent maintainability prediction models. |
Year | DOI | Venue |
|---|---|---|
2020 | 10.1109/CEC48606.2020.9185809 | CEC |
DocType | Citations | PageRank |
Conference | 0 | 0.34 |
References | Authors | |
0 | 2 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Ruchika Malhotra | 1 | 533 | 35.12 |
| Kusum Lata | 2 | 0 | 0.34 |