Anzeige
Erschienen in:
02.05.2018 | Correction
Correction to: Streaming active learning strategies for real-life credit detection: assessment and visualization
Erschienen in: International Journal of Data Science and Analytics | Ausgabe 4/2018
EinloggenAktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
Excerpt
A line was misplaced in Table 1. The correct Table 1 is given below. The original article has been corrected.
Table 1
Summary of active learning and semi-supervised strategies described in the paper
|
Id
|
Strategy
|
Type
|
|---|---|---|
|
HRQ
|
Highest risk querying
|
Baseline/BL (Sect. 5.1)
|
|
R
|
Random querying
|
Exploratory active learning/EAL (Sect. 5.2)
|
|
P
|
Unsupervised (PCA) querying
|
|
|
U
|
Uncertainty querying
|
|
|
M
|
Mix of random and uncertainty querying
|
|
|
SR
|
SSSL on random points
|
Stochastic semi-supervised learning/SSSL (Sect. 5.3)
|
|
SU
|
SSSL on uncertain points
|
|
|
SM
|
SSSL on random/uncertain points
|
|
|
SE
|
SSSL on points most likely to be genuine
|
|
|
SR-U
|
SSSL on uncertain points \(+\) random sampling
|
SSSL \(+\) EAL (Sect. 5.3)
|
|
SR-R
|
SSSL on random points \(+\) random sampling
|
|
|
SR-M
|
SSSL on random/uncertain points \(+\) random sampling
|
|
|
SRN[p]
|
SR with reduced x% of negative feedback
|
Modified SSSL (Sect. 5.3)
|
|
ROS
|
Random oversample
|
Oversample (Sect. 5.4)
|
|
SMOTE
|
SMOTE
|
|
|
QFU
|
Querying by frequent uncertainty
|
Multiple instance learning (Sect. 5.5)
|
|
MF-...
|
Max combining function
|
|
|
SM-...
|
Softmax combining function
|
|
|
LF-...
|
Logarithmic combining function
|