Cover Image
QR Code

Learning More About Active Learning.

The article discusses improvements and innovation of the field of active learning algorithms that are creating large savings in label complexity. In an attempt to prevent junk email from reaching its intended recipients, early spam filters relied on a combination of brute force computing with passiv...

Full description

Published in:Communications of the ACM Vol. 52; no. 4; pp. 11 - 14
Main Author: Stemp-Morlock, Graeme
Format: Article
Published: Association for Computing Machinery Apr2009
Subjects:
Spam filtering (Email)
Spam email
Algorithm research
Filters (Mathematics)
Email security
Hanneke, Steve
University faculty
Students
Online Access:View this record in EBSCOhost
Summary:The article discusses improvements and innovation of the field of active learning algorithms that are creating large savings in label complexity. In an attempt to prevent junk email from reaching its intended recipients, early spam filters relied on a combination of brute force computing with passive learning and refined processing with active learning, the article indicates. Commentary is provided by Ph.D. candidate Steve Hanneke on improvements made to active learning processes and informative examples as well as activized learning.