1. ROC Curves
In Receiver Operating Characteristic (ROC) curve the true positive rate (Sensitivity) is plotted in function of the false positive rate (100-Specificity) for different cut-off points for each classifier applied to our Dataset.
Each point on the ROC curve represents a sensitivity/specificity pair corresponding to a particular decision threshold.
2. Graphs Plotted
The overall accuracy compares all four ROC curves for respective classifiers in one Graph .
Individual ROC curves for each classifier are plotted along with the zoomed X and Y labels regions give a comprehensive view of the accuracy pattern of each classifier algorithm .
All the Graphs are plotted using Google Chart API and Matlab.
3. Result Analysis
Total Number of instances to be classified = 4493
Test Mode for each classifier below : 10-fold cross-validation
Hence we can conclude that by including a feature namely, number of entries in the import address table, quantitatively selected by information gain values of each feature, we could significantly improve the classification of packed executables with lower false negatives and low computational complexity.