Graphics Reference
In-Depth Information
Table 5.1
Filtering approaches by category as of [
19
]
Detection based on thresholding of a measure
Partition filtering for large data sets
Measure: classification confidence
[
82
]
For large and distributed data sets
[
102
,
103
]
Least complex correct hypothesis
[
24
]
Model influence
Classifier predictions based
LOOPC
[
57
]
Cost sensitive learning based
[
101
]
Single perceptron perturbation
[
33
]
SVM based
[
86
]
Nearest neighbor based
ANN based
[
42
]
CNN
[
30
]
Multi classifier system
[
65
]
BBNR
[
15
]
C4.5
[
44
]
IB3
[
3
]
Nearest instances to a candidate
[
78
,
79
]
Tomek links
[
88
]
Voting filtering
PRISM
[
81
]
Ensembles
[
10
,
11
] DROP
[
93
]
Bagging
[
89
]
Graph connections based
ORBoost
[
45
]
Grabiel graphs
[
18
]
Edge analysis
[
92
]
Neighborhood graphs
[
66
]
sensitive to class noise as well [
74
]. This instability has make them very suitable for
ensemble methods. As a countermeasure for this lack of stability some strategies can
be used. The first one is to carefully select an appropriate splitting criteria measure.
In [
2
] several measures are compared to minimize the impact of label noise in the
constructed trees, empirically showing that the imprecise info-gain measure is able
to improve the accuracy and reduce the tree growing size produced by the noise.
Another approach typically described as useful to deal with noise in decision trees
is the use of pruning. Pruning tries to stop the overfitting caused by the overspecial-
ization over the isolated (and usually noisy) examples. The work of [
1
] eventually
shows that the usage of pruning helps to reduce the effect and impact of the noise in
the modeled trees. C4.5 is the most famous decision tree and it includes this pruning
strategy by default, and can be easily adapted to split under the desired criteria.
We have seen that the usage of ensembles is a good strategy to create accurate
and robust filters. Whether an ensemble of classifiers is robust or not against noise
can be also asked.
Many ensemble approaches exist and their noise robustness has been tested. An
ensemble is a system where the base learners are all of the same type built to be as
varied as possible. The two most classic approaches bagging and boosting were com-
pared in [
16
] showing that bagging obtains better performance than boosting when
label noise is present. The reason shown in [
1
] indicates that boosting (or the par-
ticular implementation made by AdaBoost) increase the weights of noisy instances
too much, making the model construction inefficient and imprecise, whereas mis-
labeled instances favour the variability of the base classifiers in bagging [
19
]. As
AdaBoost is not the only boosting algorithm, other implementations as LogitBoost
and BrownBoost have been checked as more robust to class noise [
64
]. When the base
