An Informative and Comprehensive Behavioral Characteristics Analysis Methodology of Android Application for Data Security in Brain-Machine Interfacing

Table 3

Performance of single behavioral characteristic.

Behavioral characteristic	Algorithm	TP	FP	Precision	Recall	F-measure	ROC area
Code	SVM	0.685	0.318	0.686	0.685	0.685	0.684
	KNN	0.692	0.324	0.692	0.692	0.689	0.734
	Random forest	0.694	0.323	0.694	0.694	0.69	0.735
Hardware	SVM	0.548	0.535	0.563	0.548	0.411	0.507
	KNN	0.548	0.534	0.564	0.548	0.413	0.543
	Random forest	0.548	0.534	0.564	0.548	0.413	0.547
Indent	SVM	0.689	0.489	0.762	0.689	0.723	0.69
	KNN	0.836	0.184	0.849	0.836	0.833	0.885
	Random forest	0.84	0.181	0.854	0.84	0.837	0.892
Requested permission	SVM	0.873	0.137	0.876	0.873	0.872	0.868
	KNN	0.936	0.064	0.936	0.939	0.936	0.972
	Random forest	0.937	0.063	0.938	0.937	0.937	0.975
Suspicious API	SVM	0.753	0.0.257	0.752	0.753	0.752	0.749
	KNN	0.782	0.228	0.782	0.782	0.781	0.861
	Random forest	0.779	0.231	0.779	0.779	0.777	0.861
Used permission	SVM	0.837	0.177	0.841	0.837	0.835	0.83
	KNN	0.864	0.138	0.864	0.864	0.864	0.942
	Random forest	0.867	0.133	0.867	0.867	0.867	0.943
API	SVM	0.843	0.125	0.859	0.843	0.851	0.905
	KNN	0.831	0.134	0.895	0.831	0.862	0.919
	Random forest	0.904	0.092	0.916	0.904	0.91	0.927
Payload	SVM	0.555	0.531	0.755	0.555	0.407	0.512
	KNN	0.699	0.313	0.698	0.699	0.697	0.774
	Random forest	0.7	0.311	0.699	0.7	0.698	0.779
Cert	SVM	0.744	0.344	0.796	0.744	0.769	0.8
	KNN	0.764	0.319	0.758	0.764	0.761	0.786
	Random forest	0.807	0.183	0.806	0.807	0.807	0.834
String	SVM	0.724	0.269	0.731	0.724	0.727	0.76
	KNN	0.544	0.544	0.544	0.544	0.544	0.544
	Random forest	0.763	0.32	0.736	0.762	0.763	0.784
Component	SVM	0.839	0.164	0.84	0.839	0.839	0.851
	KNN	0.85	0.192	0.863	0.85	0.856	0.872
	Random forest	0.893	0.117	0.893	0.847	0.87	0.875