Research Article
Prediction of Defective Software Modules Using Class Imbalance Learning
Table 2
Details of software metrics.
| Type | Metrics | Definition |
| McCabe | LOC | Number of code lines | v(g) | Cyclomatic complexity | ev(g) | Essential complexity | iv(g) | Design complexity |
| Derived Halstead | n | Number of operators and operands | v | Volume | l | Program length | d | Difficulty | i | Intelligence | e | Effort to write code | b | Effort estimate | t | Time estimator |
| Basic Halstead | lOCode | Line count | lOComment | Comment count | lOBlank | Blank line count | lOCodeAndComment | Count of code and comment lines | uniq_op | Number of unique operators | uniq_opnd | Number of unique operands | total_op | Number of total operators | total_opnd | Number of total operands | branchcount | Number of branch counts |
| Class | Defect | Describing whether a software module is defective or not |
|
|