, these new candidates become ݖଵ, ݖଶ and ݖଷ as shown in Figure

here two of them can reach the global minimum ݕଵ by using the

s method.

enetic algorithms are normally used for solving a combinational

ion problem, in which the candidates can take several limited

f status such as the binary status. The binary status includes the

and the absence for an object or a variable. If the problem is

ly large, exhausting all possible combinations in an even modern

may become infeasible. For instance, sequence homology

t, which has been introduced in Chapter 7 of this book, is such a

ional optimisation process. Fortunately, the genetic algorithms

n used to reach some optimal homology alignments between

s [Chowdhury and Garai, 2017]. The genetic algorithms have

n used for the cancer trend prediction, which was organised as a

mbinational optimisation problem [Kim, et al., 2021].

enetic programming algorithms are normally used for modelling

cated system, where a system can be expressed by a mathematical

of variables and the variables can interplay in such a system

992; Suganuma, et al., 2020]. Given a data set, the genetic

ming can be used to discover an optimal mathematical equation

h the data are assumed to be generated. Therefore, this kind of

learning algorithms can be used to discover the intelligence

n data and express the discovered intelligence using human-

rules, such as the equations.

tic programming

etic programming algorithm (GP) has had applications in the

of various biological/medical data for pattern analysis [Gao and

001; McKay, et al., 1997; Lin, et al., 2002; Yang, et al., 2003;

al., 2009]. In a GP model, a chromosome is used to express a tree.

uch a model, an algebraic equation is commonly used to represent

onship between variables. This kind of human-thinking-alike

ce is very useful in many applications, where the understanding