A Comparison of the Classification of Disparate Malware Collected in Different Time Periods

Abstract

<p class="xmsoplaintext">It has been argued that an anti-virus strategy based on malware collected at a certain date, will not work at a later date because malware evolves rapidly and an anti-virus engine is then faced with a completely new type of executable not as amenable to detection as the first was.</p><p class="AbstactText">In this paper, we test this idea by collecting two sets of malware, the first from 2002 to 2007, the second from 2009 to 2010 to determine how well the anti-virus strategy we developed based on the earlier set [18] will do on the later set. This anti-virus strategy integrates dynamic and static features extracted from the executables to classify malware by distinguishing between families. We also perform another test, to investigate the same idea whereby we accumulate all the malware executables in the old and new dataset, separately, and apply a malware versus cleanware classification.</p><p class="Abstract">The resulting classification accuracies are very close for both datasets, with a difference of approximately 5.4% for both experiments, the older malware being more accurately classified than the newer malware. This leads us to conjecture that current anti-virus strategies can indeed be modified to deal effectively with new malware.</p

Large Iterative Multitier Ensemble Classifiers for Security of Big Data

“…In [24] the authors compare two disparate sets of malware collected in different time periods and train an antivirus strategy on the earlier set to determine how well it will manage to handle the later set. An anti-virus strategy is used that integrates dynamic and static features extracted from the executables to to detect malware versus cleanware and to classify malware by distinguishing between malware families.…”

Multilayer hybrid strategy for phishing email zero‐day filtering

“…Our MHS has more layers and tackles the new problem of filtering phishing emails during a new time span utilizing only the training data of a separate previous time span. This is inspired by the previous work in the different area of malware detection, where analogous task was investigated in . In the case of phishing email, this problem can be regarded as an advanced special case of zero‐day phishing email filtering.…”