Machine Learning et Sécurité

Abstract

The field of cybersecurity is a very broad field that aims essentially at providing protection to computerized systems. Machine Learning is one of the techniques used to assist decision-making in the face of various IT security issues. The key objective of using Machine Learning is to provide solutions to issues such as authentication via biometrics, intrusion detection, botnet detection, DNS tunnel detection, detection DGAs, SQL injection detection, spam detection and alert correlation. Based on datasets, we proposed an implementation enhancement for spam detection using the Pipeline to combine the GridSearchCV to find the hyper-parameters of the Naïve Bayes. This improvement resulted in precision of 99% and recall of 99% . We will also explore other Machine Learning algorithms such as the Support Vector Machine (SVM), the decision tree, the Logistic Regression (RL) to analyze the results.

Date of Award	20 Jun 2019
Original language	French
Awarding Institution	University of Namur
Supervisor	Jean-Noel Colin (Supervisor)