Graphics Reference
In-Depth Information
Table 10.1
Statistical procedures available in KEEL
Procedure
References Description
5x2cv-f test
[
25
]
Approximate f statistical test for 5x2-CV
Ttest
[
41
]
Statistical test based on the Student's t distribution
Ftest
[
42
]
Statistical test based on the Snedecor's F distribution
Shapiro-Wilk test
[
43
]
Variance test for normality
Mann-Whitney U test
[
44
]
U statistical test of difference of means
Wilcoxon test
[
45
]
Nonparametric pairwise statistical test
Friedman test
[
46
]
Nonparametric multiple comparisons statistical test
Iman-Davenport test
[
47
]
Derivation from the Friedman's statistic (less conservative)
Bonferroni-Dunn test
[
48
]
Post-Hoc procedure similar to Dunnet's test for ANOVA
Holm test
[
49
]
Post-Hoc sequential procedure (most significant first)
Hochberg test
[
50
]
Post-Hoc sequential procedure (less significant first)
Nemenyi test
[
51
]
Comparison with all possible pairs
Hommel test
[
52
]
Comparison with all possible pairs (less conservative)
Table
10.1
shows the procedures existing in the KEEL statistical package. For
each test, a reference and a brief description is given (an extended description can
be found in the
Statistical Inference in Computational Intelligence and Data Mining
website and in the KEEL website
18
).
10.5.1 Case Study
In this section, we present a case study as an example of the functionality and process
of creating an experiment with the KEEL software tool. This experimental study
is focused on the comparison between the new algorithm imported (SGERD) and
several evolutionary rule-based algorithms, and employs a set of supervised classi-
fication domains available in KEEL-dataset. Several statistical procedures available
in the KEEL software tool will be employed to contrast the results obtained.
10.5.1.1 Algorithms and Classification Problems
Five representative evolutionary rule learningmethods have been selected to carry out
the experimental study: Ant-Miner, CO-Evolutionary Rule Extractor (CORE), HIer-
archical DEcision Rules (HIDER), Steady-State Genetic Algorithm for Extracting
Fuzzy Classification Rules From Data (SGERD) and Tree Analysis with Randomly
Generated and Evolved Trees (TARGET) methodology. Table
10.2
shows their ref-
erences and gives a brief description of each one.
