Hunting for malware with machine learning.

[kenjoe41]

Comprehensive malware research can be a difficult task. Before reversing and constructing the timeline, the reverser needs a significant set of samples of the malware from multiple stages of its development. Finding similar samples can be quite difficult, as comparing files at scale is computationally expensive and often unfruitful. Naturally, having a problem with scale and malware, we brought out the big guns: machine learning.

Read up: http://ezl.ink/t

[evolut1o]

Nice post, o'reilly have some Machine Learning books too, one edition especially for hacking, would be a total scrub if i didn't add that.

[kenjoe41]

I never realised or ever read that book. Have a link to it? Amazon or o'reilly link.

[evolut1o]

http://shop.oreilly.com/category/get/machine-learning-kit.do -> All machine learning books
http://shop.oreilly.com/product/0636920018483.do -> Hacking edition

http://www.allitebooks.com/machine-learning-for-hackers/ -> The free pdf online, don't know if the forum already have it, but in case it don't, there it is.

[Psycho_Coder]

Nice article! I would ask you to have a look at this: https://github.com/CamDavidsonPilon/Probabilistic-Programming-and-Bayesian-Methods-for-Hackers. Also if you're serious about learning ML then go through the MOOC for ML on Coursera by Andrew Ng.

I have been thinking about this in particular lately, where we apply ML or Fuzzy Logic or maybe some NFS into Malware forensics and detection of other forms of Security Threats. I have been studying about malware recently for some job related work. This is definitely a field where we could apply such technology and take it to a step further. In fall 2016 I will pursue my M.S. and I am thinking to work on this topic itself but with focus on Android.

if you find some more interesting on this, do PM me

Thanks