Requirements:
Pylibol, scipy == 1.1.0, numpy ==1.15.2
To use:
STEP 1:
python feature_extraction.py
Extract all detection features from applications
STEP 2:
python feature_set_initialization.py
Initialize the feature set according to initialization dataset
STEP 3:
python vector_generation.py
Generate feature vector for the initialization dataset
STEP 4:
python model_pool_construction.py
Initialize the mode pool using the feature vector of the initialization feature vector
STEP 5:
python classification_evolvement.py --past p --current c --starting s --low l --high h --buffer b
p = directory of previous data
c = directory of current data
s = directory of initialization data
l = low threshold value
h = high threshold value
b = buffer size