runexperiment/MLCCI

MLCCI code

The experiment is divided into two parts, one is MCLLI approach and the other is Fuzzy Weighted KNN approach which is baseline.

Running MCLLI approach experimental code

1. Executing the main function of featureExtract.py, obtain the results of four feature calculations.

2. Executing the main function of Me2.py, calculating CC recognition results, including Recall, Precision, FPR and F1-score. In addition, obtaining the random forest's training model.

3. Executing the main function in Location.py to calculate the list of suspicious statements.

4. Executing the main function in FaultMe.py to get the metrics of fault localization which contains Wasted Effort and Accuracy@N.

Running Fuzzy Weighted KNN approach experimental code

1. Executing the main function of featureExtract.py, obtain the results of four feature calculations, replace the execution in the main function with execution_baseline.

2. Executing the main function of MyTest2.py, calculating CC recognition results, including Recall, Precision, FPR and F1-score.

3. Executing the main function in base_Location.py to calculate the list of suspicious statements.

4. Executing the main function in bsse_Fault.py to get the metrics of fault localization which contains Wasted Effort and Accuracy@N.