Information Technology Reference
In-Depth Information
without RNCs performs slightly better than the ADF-RNC algorithm (aver-
age best-of-run fitnesses of 339.19 in the absence of RNCs and 339.06 in the
presence of RNCs), corroborating the results obtained in section 5.6.4 that
RNCs are not important to make an accurate diagnosis of breast cancer.
The comparison of the performances of the acellular systems with the
cellular shows that adaptation in the cellular systems occurs more slowly
than in the simpler systems without ADFs (compare Tables 5.9 and 6.7). But,
obviously, with more time or larger populations, good solutions can also be
found with the cellular systems as adaptation in them still occurs smoothly.
Consider, for instance, one of the best solutions designed with the simpler
cellular system without RNCs. It was found in generation 890 of run 75 (the
active cell is shown in bold):
Normal Genes:
+.*.d3.d8.+.-.+.d0.d3.d6.d0.d2.d3.d7.d6
+.d5.d6.+.d6.-.+.d0.d3.d1.d0.d2.d1.d7.d2
/.-.*.-.+.*.d0.d0.d4.d1.d0.d4.d3.d3.d6
Homeotic Genes:
*.-.+.+.2.-.+.0.2.2.2.1.1.1.1
+.0./.+.+.+.+.1.1.0.1.0.2.0.2
1.1.+.*./.*.2.0.0.2.2.2.2.1.0 (6.8a)
This model classifies correctly 341 out of 350 fitness cases in the training set
and 172 out of 174 sample cases in the testing set, which corresponds to a
training set classification error of 2.571% and a classification accuracy of
97.429%, and a testing set classification error of 1.149% and a classification
accuracy of 98.851%. More formally, the model (6.8a) can be translated into
the following C++ program:
double ADF0(double d[])
{
double dblTemp = 0.0;
dblTemp = ((d[8]*((d[0]-d[3])+(d[6]+d[0])))+d[3]);
return dblTemp;
}
double ADF1(double d[])
{
double dblTemp = 0.0;
dblTemp = (d[5]+d[6]);
return dblTemp;
}
Search WWH ::




Custom Search