AAAI Publications, The Thirty-First International Flairs Conference

Font Size: 
Learning to Identify Known and Unknown Classes: A Case Study in Open World Malware Classification
Mehadi Hassen, Philip K. Chan

Last modified: 2018-05-10

Abstract


In this paper we propose an open world malware classification. Our approach is not only able to identify known families of malware but is also able to distinguish them from malware families that were never seen before. Our proposed approach is more accurate and scales better on two evaluation datasets when compared to existing algorithms.

Keywords


open world classification, open set recognition, malware classification

Full Text: PDF