Graphics Reference
In-Depth Information
Table 10.2
Algorithms tested in the experimental study
Method
Reference
Description
Ant-Miner
[
53
]
An Ant Colony System based using a heuristic function based
In the entropy measure for each attribute-value
CORE
[
54
]
A coevolutionary method which employs as fitness measure a
Combination of the true positive rate and the false positive rate
HIDER
[
55
]
A method which iteratively creates rules that cover
Randomly selected examples of the training set
SGERD
[
56
]
A steady-state GA which generates a prespecified number
Of rules per class following a GCCL approach
TARGET
[
57
]
A GA where each chromosome represents a complete decision tree
On the other hand, we have used 24 well-known classification data sets (they are
publicly available on the KEEL-dataset repository web page,
19
including general
information about them, partitions and so on) in order to check the performance of
these methods. Table
10.3
shows their main characteristics where #
Ats
is the number
of attributes, #
Ins
is the number of instances and #
Cla
is the number of Classes. For
each data set the number of examples, attributes and classes of the problem described
are shown. We have employed a 10-FCV procedure as a validation scheme to perform
the experiments.
Table 10.3
Data sets employed in the experimental study
Name
#Ats
#Ins
#Cla
Name
#Ats
#Ins
#Cla
HAB
3
306
2
Wisconsin
9
699
2
IRI
4
150
3
Tic-tac-toe
9
958
2
BAL
4
625
3
Wine
13
178
3
NTH
5
215
3
Cleveland
13
303
5
MAM
5
961
2
Housevotes
16
435
2
BUP
6
345
2
Lymphography
18
148
4
MON
6
432
2
Ve h i c l e
18
846
4
CAR
6
1,728
4
Bands
19
539
2
ECO
7
336
8
German
20
1,000
2
LED
7
500
10
Automobile
25
205
6
PIM
8
768
2
Dermatology
34
366
6
GLA
9
214
7
Sonar
60
208
2