Classification Of Malware Families Using Naïve Bayes Classifier

Due to the increase in Android smartphone users, it is directly proportional to the rapid
increase in malware development. It is not uncommon for research on malware every year to
discuss malware families with various approaches, one of which is machine learning. By getting
credible malware data, it can make it easier for researchers to analyze malware. There is a
publicly accessible malware database created by the Canadian Institute for Cybersecurity (CIC).
This data is called CICInvestAndMal2019 which contains malware data. This dataset is created
by performing static and dynamic analysis on a smartphone in real time. The results of the
analysis were then processed using the Random Forest method which resulted in a precision of
61.2% and a recall of 57.7%. Based on this research, the author will classify the
CICInvestAndMal2019 dataset using the Naïve Bayes method, and the results obtained from the
Naïve Bayes classification are recall and precision values of 68% and 66%, respectively.