Self-adaptive parameter and strategy based particle swarm optimization for large-scale feature selection problems with multiple classifiers

Y Xue, T Tang, W Pang, AX Liu - Applied Soft Computing, 2020 - Elsevier
… to adaptively adjust their candidate solution generation … In SABC-SI, a novel population
initialization method based on … is given in Table 2, where θ is a threshold used to determine

Adaptive threshold optimisation for online feature selection using dynamic particle swarm optimisation in determining feature relevancy and redundancy

EAK Zaman, A Ahmad, A Mohamed - Applied Soft Computing, 2024 - Elsevier
… This section describes a novel approach for OFS that leverages the power of Dynamic PSO
… In this step, the algorithm applies the PSO to the Candidate Features (CF) in order to obtain …

A novel intelligent diagnosis method using optimal LS-SVM with improved PSO algorithm

W Deng, R Yao, H Zhao, X Yang, G Li - Soft computing, 2019 - Springer
… a self-adaptive artificial bee colony algorithm based on the global best candidate for solving
… The particle dimension of the improved PSO algorithm is determined, and the position and …

[HTML][HTML] Novel mode adaptive artificial neural network for dynamic learning: Application in renewable energy sources power generation prediction

MA Zamee, D Won - Energies, 2020 - mdpi.com
… -based algorithms: Advanced Particle Swarm Optimization (APSO), … In this work, to calculate
the error during the learning process… where x k , i , g is the i t h candidate’s value of the k t h …

A novel hybrid wrapper–filter approach based on genetic algorithm, particle swarm optimization for feature subset selection

F Moslehi, A Haeri - Journal of Ambient Intelligence and Humanized …, 2020 - Springer
… Each string is the encoded binary, real, etc. version of a candidate solution. An evaluation …
After generating the initial population and calculating the cost of each member, the method is …

Efficient feature selection algorithm based on particle swarm optimization with learning memory

B Wei, W Zhang, X Xia, Y Zhang, F Yu, Z Zhu - IEEE Access, 2019 - ieeexplore.ieee.org
… , a novel feature selection algorithm based on PSO with … λ as a threshold to determine
whether a feature is selected or … When evaluating a candidate feature subset, we can check …

An adaptive hybrid differential evolution algorithm for continuous optimization and classification problems

HT Rauf, WHK Bangyal, MI Lali - Neural Computing and Applications, 2021 - Springer
… to determine the mutant vector for the mutation operation. Similarly, for CMHDE-PSO (1)
novel … strategies on each portion of the populace to broaden the candidate vector solution. …

Novel self-adjusted particle swarm optimization algorithm for feature selection

B Wei, X Wang, X Xia, M Jiang, Z Ding, Y Huang - Computing, 2021 - Springer
… in classification problems, two major factors determine the … Before evaluating a candidate
feature subset, we can check whether it has … The detail of evaluating a candidate feature subset …

Adaptive dynamic meta-heuristics for feature selection and classification in diagnostic accuracy of transformer faults

SSM Ghoneim, TA Farrag, AA Rashed… - Ieee …, 2021 - ieeexplore.ieee.org
… ; then, the PSO-FS optimizer is used to determine the optimal … not also good candidates for
the classification process. Figure … This paper proposed a novel Adaptive Dynamic Polar Rose …

Multi-population adaptive genetic algorithm for selection of microarray biomarkers

AK Shukla - Neural Computing and Applications, 2020 - Springer
… a novel hybrid feature selection approach with a combination of particle swarm optimization
with … ; also, it is used for the selection of one or more combinations of candidate biomarkers. …