Machine Learning Approach to Detect Malicious Mobile Apps - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Communication Dans Un Congrès Année : 2022

Machine Learning Approach to Detect Malicious Mobile Apps

Hassan Kazemian
  • Fonction : Auteur
  • PersonId : 1014068

Résumé

Malicious developers are developing unsafe mobile apps which puts users at risk of exposing their personal data in unsafe hands. They are using techniques that change over time and their intention is to bypass the detector systems which are mostly rule-based. This paper avoids the limitations of rule-based systems by building a novel malware detector that can detect malicious apps by making use of machine learning techniques primarily focusing on deep neural networks i.e. deep multi-layer perceptron. These techniques have various properties that can adapt and identify various types of malicious applications. Simulation results on various datasets demonstrate clear superiority of this detector over other approaches, as this approach achieves 99% accuracy. Also, the detector is efficient enough to detect within 100 ms or less due to the intelligent use of autoencoder which reduces the dimensions in the feature.
Fichier sous embargo
Fichier sous embargo
0 4 6
Année Mois Jours
Avant la publication
mercredi 1 janvier 2025
Fichier sous embargo
mercredi 1 janvier 2025
Connectez-vous pour demander l'accès au fichier

Dates et versions

hal-04668661 , version 1 (07-08-2024)

Licence

Identifiants

Citer

Hassan Kazemian. Machine Learning Approach to Detect Malicious Mobile Apps. 18th IFIP International Conference on Artificial Intelligence Applications and Innovations (AIAI), Jun 2022, Hersonissos, Greece. pp.124-135, ⟨10.1007/978-3-031-08337-2_11⟩. ⟨hal-04668661⟩
0 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Mastodon Facebook X LinkedIn More