The 18th European Conference on Machine Learning (ECML) and the 11th European Conference on Principles and Practice of Knowledge Discovery in Databases (PKDD)

Fighting Spam under Attack: Some Notes from the Field

author: Aleksander Kolcz , Microsoft Research

You might be experiencing some problems with Your Video player.

*Slides*
0:00	Filtering Spam Under Attack: some notes from the field
1:22	Graph - Source: Pew Internet & American Life Project
2:04	Email addiction
2:28	Too bad we have spam
3:31	Overview
4:13	Email : some interesting problems
4:58	Other interesting email research
6:06	Solutions to Spam
6:20	Machine Learning/Text Mining
7:35	What is spam ? (1)
9:32	What is spam ? (2)
11:11	What is spam ? (3)
11:55	Spam campaign randomization
12:41	What is spam ? (4)
13:24	What is spam ? (5)
14:51	Examples of good-mail
17:27	Label noise
18:56	Operating point worries
20:38	The cost formula and its problems (1)
22:48	The cost formula and its problems (2)
23:57	Adversarial attacks and why size matters
25:15	Adversarial attacks: agility counts
26:36	What Happened When we Shipped an Adaptive Spam Filter
27:20	Threshold Drift Conservative Threshold Setting
27:54	Threshold Drift Lots of Spam Classified as Good
28:35	Threshold Drift (1)
28:55	Threshold Drift (2)
29:02	Adaptation with partial user feedback is hard
29:48	Attack vectors/techniques
31:41	The Hitchhiker Chaffer
32:48	Hitchhiker Chaffer’s Later Work: invisible ink

Lecture rating

People found this lecture:
Worth seeing
because it is:
Valuable and informative
Well presented
Easily understandable
Acceptably recorded
You need to login to cast your vote.

Report a problem or upload files

If you have found a problem with this lecture or would like to send us extra material, articles, exercises, etc., please use our ticket system to describe your request and upload the data.
Enter your e-mail into the 'Cc' field, and we will keep you updated with your request's status.