Skip to content

KMKnation/Malware-Detection-of-Android-Apps

Repository files navigation

Malware Detection

Step 1: Create two seperate vitual enviorment of python3.5 and python2.7 as venv and venv2 respectively.

Step 2: Run runme.sh in venv of python3.5 with following command in terminal.

./runme.sh

Step 3: Extract the static data from DATASET which generated the data.json by executing following command with python3 venv

 python prepare_dataset.py

Extract Features

Step 4: Run runme2.sh for installing python2 dependencies.

./runme2.sh

Step 5: Check VT_ANALYSIS folder. is there any pre-analysis file of virus total exists or not. If Exists then skip this step.

python2 virustotal_analyser.py

Dynamic Analysis

Step 6: Run the following script from python3's venv to generate feature vector and training the classifier.

python jsontovector.py

Output:

Classifier Accuracy
SVC Score 0.9318181818181818
AdaBoostClassifier Score 1.0
DecisionTreeClassifier Score 0.9545454545454546
VotingClassifier Score 0.9659090909090909