Energy and Cost Aware Workflow Offloading Using Quantum Inspired Differential Evolution in the Cloud Environments

Journal of Network and Systems Management, Journal Year: 2024, Volume and Issue: 33(1)

Published: Dec. 25, 2024

Language: Английский

Multiple Sclerosis Recognition by Wavelet Entropy and Self-adaptive PSO

Published: Nov. 13, 2023

Multiple sclerosis is a chronic, autoimmune disease that mainly affects the central nervous system, including brain, spinal cord, and optic nerve. This can cause clinical symptoms such as cognitive decline, muscle weakness, spasms, fatigue in patients, onset tends to be younger. Current medication only prevent or alleviate symptoms, so early diagnosis of this increase patients' chances treatment. Although use nuclear magnetic resonance detection improve efficiency auxiliary diagnosis, it still requires experienced doctors spend too much time energy on comprehensive judgment. To reduce cost multiple article proposes recognition algorithm for based wavelet entropy self-adaptive particle swarm optimization. Firstly, triple discrete transform performed brain image sclerosis, then 10 entropies are extracted from decomposed subbands, which feature dimensions image; Then, optimization used optimize feedforward neural network, order obtain optimal connection weights thresholds during training process. The result model with an average sensitivity 92.29±1.89, specificity 92.54±0.67, precision 92.48±0.59, accuracy 92.42±0.88, F1 score 84.85±1.74, Matthews correlation coefficient 92.37±0.96, Fowlkes223 Mallows Index 92.38±0.96. experimental results indicate has very important data support role detecting sclerosis.

Language: Английский

Competitive Swarm Optimizer Based on Individual Learning Mechanism

Published: Dec. 6, 2023

In view of the disadvantages traditional competitive swarm optimizer (CSO), such as falling into local minimization or poor convergence accuracy, this paper proposed an enhanced CSO algorithm called based on individual learning mechanism (ILCSO). Firstly, selection rate and are designed to dynamically select winner loser. The losers updated by precise strategy improve exploitation ability. Secondly, mutation performance improvement is introduced, which improves search ability algorithm. effectively balances global exploration with mechanism, probability finding optimal solution. Finally, ILCSO compared six classical meta-heuristic algorithms CEC2014 benchmark functions. Wilcoxon rank-sum test used demonstrate that effective. Experimental results statistical analysis show has higher speed accuracy.

Language: Английский