A review of soft computing technology applications in several mining problems

Soft computing-based localizations in wireless sensor networks

“…In general, given a set of training data, each has been mapped into two classes by a SVM, which makes it a non-probabilistic binary linear classifier [17,18]. SVM is modeled as a representation of points in space that are mapped for the purpose of separating the category that has the highest margin, as illustrated in Fig.…”

“…The Fuzzy Logic System (FLS) is used to manage reasoning in which there is an approximation that does not provide the exact solution, and in general, the true value will be given in a range of 0 and 1 [17,18]. Fig.…”

“…In general, Genetic Algorithms (GAs) are one of the evolutionary computations aside from Particle Swarm Optimization (PSO) and DE [17,18]. There are three main steps in GAs, as follows: ) -• Population selection: this stage is used to properly select the population members with a size constraint imposed on the parents for the current generation.…”

“…In general, soft computing techniques are used to resolve the problem with an uncertainty solver as well as a non-linear solver [17,18]. There are several classes of soft computing, e.g., Neural Networks (NN), Fuzzy Logic (FL), Support Vector Machine (SVM), and Evolutionary Computation (EC) [50][51][52].…”

“…The Neural Network (NN) is one of the mathematical models that is used to imitate the human brain in performing an intelligent task [17,18]; it integrates computational units (neurons) in multilayers in which the layers are connected by adjustable weights. Traditionally, there are three layers: Input, Hidden, and Output.…”