In the Internet age, malicious software (malware) represents a serious threat to the security of information systems. Malware-detection systems to protect computers must perform a real-time analysis of the executable files. The paper shows that machine-learning methods can support the challenging, yet critical, task of unseen malware recognition, i.e., the classification of malware variants that were not included in the training set. The experimental verification involved a publicly available dataset, and confirmed the effectiveness of the overall approach.
Machine learning-based system for detecting unseen malicious software
BISIO, FEDERICA;GASTALDO, PAOLO;MEDA, CLAUDIA;NASTA, STEFANO;ZUNINO, RODOLFO
2016-01-01
Abstract
In the Internet age, malicious software (malware) represents a serious threat to the security of information systems. Malware-detection systems to protect computers must perform a real-time analysis of the executable files. The paper shows that machine-learning methods can support the challenging, yet critical, task of unseen malware recognition, i.e., the classification of malware variants that were not included in the training set. The experimental verification involved a publicly available dataset, and confirmed the effectiveness of the overall approach.File in questo prodotto:
File | Dimensione | Formato | |
---|---|---|---|
Bisio2016_Chapter_MachineLearning-BasedSystemFor.pdf
accesso chiuso
Tipologia:
Documento in versione editoriale
Dimensione
158.86 kB
Formato
Adobe PDF
|
158.86 kB | Adobe PDF | Visualizza/Apri Richiedi una copia |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.