veyepar: scikit-learn, machine learning and cybercrime attribution

Author(s):	Robert Layton
Location	Tasman A
Date	jul Sat 06	Days Raw Files
Start	13:40	First Raw Start	13:35
Duration	0:30:00	Offset	0:04:47
End	14:10	Last Raw End	14:53
Chapters	00:00
Total cuts_time	29 min.

http://2013.pycon-au.org/schedule/30019/view_talk
raw-playlist raw-mp4-playlist encoded-files-playlist host archive public mp4 svg png

assets release.pdf scikitlearn_machine_learning_and_cybercrime_attribution.json
logs
Admin: episode episode list cut list raw files day marks day marks day image_files

State:

Locked:

clear this to unlock

Locked by:

user/process that locked.

Start:

initially scheduled time from master, adjusted to match reality

Duration:

length in hh:mm:ss

Name:

Video Title (shows in video search results)

Emails:

email(s) of the presenter(s)

Released:

has someone authorised pubication

Normalise:

Channelcopy:

m=mono, 01=copy left to right, 10=right to left, 00=ignore.

Thumbnail:

filename.png

Description:

markdown

The scikit-learn library is a rapidly growing open source toolkit for machine learning in python. It allows for practitioners and researchers to apply machine learning in a variety of applications and is used by companies worldwide. Developed by programmers from around the world, the project has a large (and increasing) number of machine learning algorithms, a very useful set of utility functions and has also spawned a set of detailed tutorials. Written in python with the aid of numpy, scipy and cython, this library is featured, fast and extensible.

In this talk, I will introduce scikit-learn, giving an overview of the library, its features and how to use it for a number of different applications. Next talk about some of the tutorials that are actively being developed for learning machine learning and scikit-learn and also how to contribute. Through this, I'll introduce some key machine learning concepts and how you can apply them to a variety of tasks. The focus will be on practical uses, rather than theoretical advances.

To end the presentation, I'll briefly overview the research I perform at the Internet Commerce Security Laboratory (University of Ballarat) in cybercrime attribution, where I work with our industry partners to disrupt cybercrime. While it can be very difficult to do direct network based attribution, indirect methods through criminal profiling may assist in stopping crimes such as phishing or online fraud. I'll walk through some of our results in identifying the size and scope of the operations behind some of these attacks.

Comment:

production notes