WIDM 04: Lee et al. Co-training Web Block
Classification
17
Rough
grained model
•Slightly different model
of splitting than earlier work
•Smaller amount of training
examples
•No significant gain from co-training but comparable
to other work (19.5% error vs. 14-18 error%)
•Use only three categories:
-Related
-Important
-Unimportant
•Advocated by
human agreement, well-founded (Song et
al. 04)