Survey on the Usage of Machine Learning Techniques for Malware Analysis.

摘要: Coping with malware is getting more and challenging, given their relentless growth in complexity volume. One of the most common approaches literature using machine learning techniques, to automatically learn models patterns behind such complexity, develop technologies for keeping pace speed development novel malware. This survey aims at providing an overview on way has been used so far context analysis. We systematize surveyed papers according objectives (i.e., expected output, what analysis to), information about they specifically use features), techniques employ algorithm process input produce output). also outline a number problems concerning datasets considered works, finally introduce concept economics, regarding study existing tradeoffs among key metrics, as accuracy economical costs.