Information Technology Reference
In-Depth Information
Gene 2: *a?E-+EQQEbcaacbab??baaab1126452672558
C
2
: {66, 6, 39, 22, 7, 27, 12, 98, 85, 71}
Gene 3: +?—bL/Qb-Q*??cbbccc?bb?c7785717851803
C
3
: {79, 92, 92, 20, 74, 34, 40, 5, 21, 67}
Gene 4: -+-*?c?E*-a-??c?cc?b?bbab6113140654932
C
4
: {80, 14, 56, 18, 97, 92, 75, 33, 34, 61}
Gene 5: Q?PEaEL*Ea*L?c????bacbc?c1072195508656
C
5
: {14, 88, 68, 39, 1, 11, 80, 85, 33, 35}
Homeotic Gene: -+14/4+3/*31220320254
(6.7a)
This model has a fitness of 66.1181 and an R-square of 0.9591176321 evalu-
ated against the training set of 40 fitness cases and, therefore, is consider-
ably better than the model (6.6) designed without random numerical con-
stants. More formally, the model (6.7a) can be expressed by the following
C++ program:
double ADF
0(double d[])
{
double dblTemp = 0.0;
dblTemp = pow(45,d[2]);
return dblTemp;
}
double ADF1(double d[])
{
double dblTemp = 0.0;
dblTemp = d[0];
return dblTemp;
}
double ADF2(double d[])
{
double dblTemp = 0.0;
dblTemp = (d[0]*6);
return dblTemp;
}
double ADF3(double d[])
{
double dblTemp = 0.0;
dblTemp = (5+((log(sqrt(sqrt(21)))-(d[1]/
((d[2]*d[1])-5)))-d[1]));
return dblTemp;
}
