Research Article

A Malware Detection Scheme Based on Mining Format Information

Table 5

Experimental results for all classification algorithms (feature selection only from the training set).

Feature selection methodThe number of featuresAlgorithmTPR (%)FPR (%)Accuracy (%)AUC

Filter18J4898.71.598.60.988
Random Forest99.21.598.90.997
Adaboost (J48)99.11.598.80.999
Bagging (J48)99.11.099.10.998

Wrapper19Random Forest99.11.3990.998