Titolo:  Explaining black-box android malware detection
Autori: 
Data di pubblicazione:  2018
Autori:  Melis, Marco; Maiorca, Davide; Biggio, Battista; Giacinto, Giorgio; Roli, Fabio
Presenza coautori internazionali:  no
Lingua:  Inglese
Titolo del libro:  26th European Signal Processing Conference, EUSIPCO 2018
ISBN:  978-9-0827-9701-5
Editore:  IEEE, Institute of Electrical and Electronics Engineers
Serie:  EUROPEAN SIGNAL PROCESSING CONFERENCE
Pagina iniziale:  524
Pagina finale:  528
Numero di pagine:  5
Digital Object Identifier (DOI):  http://dx.doi.org/10.23919/EUSIPCO.2018.8553598
Codice identificativo Scopus:  2-s2.0-85059820168
Codice identificativo ISI:  WOS:000455614900106
URL:  https://ieeexplore.ieee.org/document/8553598
Revisione (peer review):  Esperti anonimi
Nome del convegno:  The 26th European Signal Processing Conference (EUSIPCO)
Periodo del convegno:  September 3-7, 2018
Luogo del convegno:  Rome, Italy
Abstract:  Machine-learning models have been recently used for detecting malicious Android applications, reporting impressive performances on benchmark datasets, even when trained only on features statically extracted from the application, such as system calls and permissions. However, recent findings have highlighted the fragility of such in-vitro evaluations with benchmark datasets, showing that very few changes to the content of Android malware may suffice to evade detection. How can we thus trust that a malware detector performing well on benchmark data will continue to do so when deployed in an operating environment? To mitigate this issue, the most popular Android malware detectors use linear, explainable machine-learning models to easily identify the most influential features contributing to each decision. In this work, we generalize this approach to any black-box machine- learning model, by leveraging a gradient-based approach to identify the most influential local features. This enables using nonlinear models to potentially increase accuracy without sacrificing interpretability of decisions. Our approach also highlights the global characteristics learned by the model to discriminate between benign and malware applications. Finally, as shown by our empirical analysis on a popular Android malware detection task, it also helps identifying potential vulnerabilities of linear and nonlinear models against adversarial manipulations.
Tipologia: 4.1 Contributo in Atti di convegno

File in questo prodotto:
File Descrizione Tipologia Licenza  
melis18-eusipco-v2.pdf  versione pre-print Open Access Visualizza/Apri
melis18-eusipco-v3.pdf  versione post-print Administrator   Richiedi una copia

Questionario e social

Condividi su: