Statistical Learning Applied to Malware Detection
Journal
Artificial Intelligence and Applied Mathematics in Engineering Problems
Lecture Notes on Data Engineering and Communications Technologies
ISSN
2367-4512
2367-4520
Date Issued
2020
Author(s)
Marmolejo Saucedo, José Antonio
Type
text::conference output::conference proceedings::conference paper
Abstract
This work shows an application of statistical learning methodologies in order to determine the important factors for malware detection. Support Vector Machines and Lasso Regression performed Malware classification with additional re-sampling methods. The results show that the Lasso Regression allows an efficient selection of relevant variables for the construction of the classifier, also the integration of support vector machines improves the efficiency of the classifier through the application of resampling methods. The model presented in this paper uses a statistical learning approach through the selection of variables, non-linear classification, and resampling methods. © 2020, Springer Nature Switzerland AG.