Software Defects Predictions using SQL Complexity and Naïve Bayes
Software defects result in unreliable software, therefore predicting software defects is an effort to produce quality software. In this study, we used the naïve bayes method because it has the appropriate characteristics of the data used. The data used include NASA MDP datasets and datasets from the...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Institut Teknologi Dirgantara Adisutjipto
2025-06-01
|
| Series: | Compiler |
| Subjects: | |
| Online Access: | https://ejournals.itda.ac.id/index.php/compiler/article/view/2979 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Software defects result in unreliable software, therefore predicting software defects is an effort to produce quality software. In this study, we used the naïve bayes method because it has the appropriate characteristics of the data used. The data used include NASA MDP datasets and datasets from the calculation of the sql complexity method on eight software modules. The use of two datasets was carried out because in the NASA MDP datasets there were no attributes that paid attention to the use of sql commands, therefore in the datasets from the eight software modules the sql complexity attribute was included which paid attention to the level of complexity of the use of sql commands in each module. The prediction results of this study were evaluated by considering the values of accuracy, precision, recall, and f-measure. Based on these results, the accuracy results of CM1 were 88%, PC2 was 97%, and KC3 was 78%. |
|---|---|
| ISSN: | 2252-3839 2549-2403 |