Sign inSign up
Using GNUsmail to Compare Data Stream Mining Methods for On-line Email Classification thumbnail
slide-image
Pause
Mute
Subtitles not available
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

Using GNUsmail to Compare Data Stream Mining Methods for On-line Email Classification

Published on Nov 11, 20113086 Views

Manuel Baena-Garcia

Real-time classification of emails is a challenging task because of its online nature, and also because email streams are subject to concept drift. Identifying email spam, where only two different lab

WAPA 2011 - Castro Urdiales

Related categories

Machine Learning

Chapter list

Using GNUsmail to Compare Data Stream Mining Methods for On-line Email Classification00:00
Content00:00
Context - Email mining00:16
Context - Email classification approaches00:55
Context - Hypothesis01:29
Context01:42
Content - GNUsmail02:51
GNUsmail: Architecture and Characteristics02:52
Text Preprocessing Module03:54
Learning Module - 104:53
Learning Module - 205:48
Content - Evaluation06:11
Evaluation of Data Stream Mining Methods06:14
Comparing the performance07:15
Content - Replicable Experimentation07:54
Experimental Setup - 107:55
Experimental Setup - 209:03
Results10:01
Results for beck-s11:22
Results for kitchen-l12:33
Adapted McNemar test - 113:39
Adapted McNemar test - 214:39
Content - Conclusion15:06
Conclusion and Future Work15:07