Information Technology Reference
In-Depth Information
1
0.9
0.8
Normal
REA
SERA
0.7
SMOTE
0.6
0.5
UB
0.4
10
20
30
40
50
60
70
80
90
100
(a) OA of algorithms in comparison
for SHP dataset under setup 1
1
0.95
REA
0.9
SERA
0.85
Normal
UB
0.8
0.75
0.7
0.65
SMOTE
10
20
30
40
50
60
70
80
90
100
(b) AUC of algorithms in comparison
for SHP dataset under setup 1
Figure 7.10 OA and AUROC for SHP dataset under setup 1 .
7.4.3 Hypothesis Removal
In the scenario of long-term learning from data streams, retaining all hypothe-
ses in memory over time may not be a good option. Besides the concern for
memory occupation, hypotheses built in the distant past may hinder the classifi-
cation performance on current testing dataset, which should therefore somehow
be pruned/removed from the hypotheses set H . This is an issue that is unique for
REA, as other algorithms either just rely on the hypothesis created on the current
Search WWH ::




Custom Search