| Titre : | Machine Learning in Malware Detection and Classification | | Type de document : | projet fin études | | Auteurs : | Mohammed REGUIBI, Auteur | | Langues : | Français (fre) | | Catégories : | BIG DATA
| | Index. décimale : | mast 277/19 | | Résumé : | Malware detection is an important factor in the security of computer systems. However, currently used signature-based methods cannot provide accurate detection of "Zero-day" attacks and polymorphic malwares. The classification of malware has become very important to know the basis of the behaviour of malwares and to fight back cybercriminals.
The purpose of this work is to determine the best methods of extraction, representation of characteristics, as well as classification methods to obtain the best accuracy. Specifically, k-nearest neighbors (KNN), decision trees (DT), support vector machines (SVM), classifiers Naive Bayes (NB), Random Forest (RF), and XGBoost (Extreme Gradient Boosting) are the algorithms considered in this work. The dataset used for this study consists of a set of 10,868 malware files from 9 families of different types.
This work presents the recommended methods for classifying and detecting malware based on machine learning, as well as the steps for its implementation. In addition, the study may be useful as a basis for further research in the field of malware analysis with machine learning methods. |
Machine Learning in Malware Detection and Classification [projet fin études] / Mohammed REGUIBI, Auteur . - [s.d.]. Langues : Français ( fre) | Catégories : | BIG DATA
| | Index. décimale : | mast 277/19 | | Résumé : | Malware detection is an important factor in the security of computer systems. However, currently used signature-based methods cannot provide accurate detection of "Zero-day" attacks and polymorphic malwares. The classification of malware has become very important to know the basis of the behaviour of malwares and to fight back cybercriminals.
The purpose of this work is to determine the best methods of extraction, representation of characteristics, as well as classification methods to obtain the best accuracy. Specifically, k-nearest neighbors (KNN), decision trees (DT), support vector machines (SVM), classifiers Naive Bayes (NB), Random Forest (RF), and XGBoost (Extreme Gradient Boosting) are the algorithms considered in this work. The dataset used for this study consists of a set of 10,868 malware files from 9 families of different types.
This work presents the recommended methods for classifying and detecting malware based on machine learning, as well as the steps for its implementation. In addition, the study may be useful as a basis for further research in the field of malware analysis with machine learning methods. |
|
Faire une suggestion Affiner la recherche
Machine Learning in Malware Detection and Classification / Mohammed REGUIBI
