Page 279 - Big Data Analytics for Intelligent Healthcare Management

11.4 LITERATURE SURVEY 273

corroborate the elevated performance of the enhanced kidney-inspired algorithm in resolving fea-
ture selection, which is an optimization problem. Ehteram et al. [26] proposed the reservoir oper-
ation by a new optimization algorithm: kidney algorithm. Their proposed article showed an
application of the kidney algorithm for reservoir operation, that utilizes three different operators:
excretion, filtration, and secretion, which produces more accurate solutions and faster convergence.
The author’s study contrasted reservoir operation optimization with KA, which is a new optimiza-
tion algorithm. The consequences illustrate that normal objective function principles and compu-
tational time for KA were all fewer than those found in the Genetic Algorithm [27],Shark
Algorithm [28], Weed Algorithm [7],Bat Algorithm [29], and Particle Swarm Optimization
[30]. In their present study, the authors applied the Borda method and demonstrated that KA
had great results, attaining the top rank as compared to the other models. Their study proved that
KA outperformed the remaining algorithms and addressed their defects to engender optimal oper-
ation regulation for decision-making aspects and reservoir systems. A wide range of real-life prob-
lems were solved with the appropriate use of MLP-ANN (Multilayer Perceptron Artificial Neural
Networks). There is good support for optimization methods in artificial neural networks for select-
ing the proper weights and accomplishing accurate outcomes. A newly enhanced optimization
method, which is a variation of the KA, can solve the problems of prediction of time series as well
as classification. Additional intensification is observed in the original KA due to the scenario that
more solutes are sieved and returned to sieved blood. In contrast, if extra solutes lead to despoil
which showcase the effective diversification. A newly developed optimization method for simulat-
ing neural networks was introduced by Jaddi et al. [31] in 2018. They described the problem of
rainfall time series prediction with the suitable balance of intensification and diversification. To
evaluate the performance of their developed method, the authors considered several standard data-
sets. Their experimental results showcase the effective performance and proved their method is
robust and can be used to solve real-life forecasting problems. Ekinci et al. [32] developed a
method for solving the tuning problem of power system stabilizers (PSS) using a recently devel-
oped population-based algorithm named KA. They used KA mainly to search for the best param-
eters. They sustained the problem of the power system by introducing an Eigen value coefficient.
The authors considered 16 machines as well as 68 bus power systems to evaluate the performance
of their developed method. Later, they compared their intended method with some highly recom-
mended and standard algorithms such as the ancient PSO (Particle Swarm Optimization) algorithm
as well as the BA (Bat Algorithm). They claimed that their experimental results outperformed the
compared methods. Homaid et al. [33] introduced a kidney algorithm for pairwise test suite gen-
eration. Pairwise testing can significantly reduce the rate of software testing and also raise the ca-
pability of fault detection. Metaheuristic algorithms have been mostly used for resolving
complicated problems of optimization as well as showing their efficacy to obtain nearly all optimal
solutions. Their study initiates a new pairwise strategy by adapting the KA. This is the first example
of adjusting the KA to produce a test suite. The author’s projected approach is known as the PKS
(Pairwise Kidney Strategy). Their study also highlights the pairwise kidney strategy design. In the
same way, they contrasted their proposed system with other detailed approaches in the literature in
the provision of test suite sizes. Finally, in their proposed method, the experiment results illustrated
that PKS (pairwise strategy) had very competitive outcomes when compared to the remaining
strategies.