Graphics Reference
In-Depth Information
Table 10.4
Parameter' values employed in the experimental study
Algorithm
Parameters
Ant-Miner
Number of ants: 3000, Maximum uncovered samples: 10, Maximum samples
by rule: 10
Maximum iterations without converge: 10
CORE
Population size: 100, Co-population size: 50, Generation limit: 100
Number of co-populations: 15, Crossover rate: 1.0
Mutation probability: 0.1, Regeneration probability: 0.5
HIDER
Population size: 100, Number of generations: 100, Mutation probability: 0.5
Cross percent: 80, Extreme mutation probability: 0.05, Prune examples
factor: 0.05
Penalty factor: 1, Error coefficient: 1
SGERD
Number of Q rules per class: Computed heuristically, Rule evaluation criteria
=2
TARGET
Probability of splitting a node: 0.5, Number of total generations for the GA:
100
Number of trees generated by crossover: 30, Number of trees generated by
mutation: 10
Number of trees generated by clonation: 5, Number of trees Generated by
immigration: 5
five learning methods used (Clas-AntMiner, Clas-SGERD, Clas-Target, Clas-Hider
and Clas-CORE).
After the models are trained, the instances of the data set are classified. These
results are the inputs for the visualization and test modules. The module Vis-Clas-
Tabular receives these results as input and generates output files with several perfor-
mance metrics computed from them, such as confusion matrices for each method,
accuracy and error percentages for each method, fold and class, and a final summary
of results. Figure
10.9
also shows another type of results flow, the node Stat-Clas-
Friedman which represents the statistical comparison, results are collected and a
statistical analysis over multiple data sets is performed by following the indications
given in [
38
].
Once the graph is defined, we can set up the associated experiment and save it as a
zip file for an off-line run. Thus, the experiment is set up as a set of XML scripts and
a JAR program for running it. Within the results directory, there will be directories
used for housing the results of each method during the run. For example, the files
allocated in the directory associated to an interval learning algorithmwill contain the
knowledge or rule base. In the case of a visualization procedure, its directory will
house the results files. The results obtained by the analyzed methods are shown in
the next section, together with the statistical analysis.
