Fitness Dependent Optimizer for IoT Healthcare using Adapted Parameters: A Case Study Implementation

Abstract

This discusses a case study on Fitness Dependent Optimizer or so-called FDO and adapting its parameters to the Internet of Things (IoT) healthcare. The reproductive way is sparked by the bee swarm and the collaborative decision-making of FDO. As opposed to the honey bee or artificial bee colony algorithms, this algorithm has no connection to them. In FDO, the search agent's position is updated using speed or velocity, but it's done differently. It creates weights based on the fitness function value of the problem, which assists lead the agents through the exploration and exploitation processes. Other algorithms are evaluated and compared to FDO as Genetic Algorithm (GA) and Particle Swarm Optimization (PSO) in the original work. The key current algorithms—The Salp-Swarm Algorithms (SSA), Dragonfly Algorithm (DA), and Whale Optimization Algorithm (WOA) have been evaluated against FDO in terms of their results. Using these FDO experimental findings, we may conclude that FDO outperforms the other techniques stated. There are two primary goals for this chapter: first, the implementation of FDO will be shown step-by-step so that readers can better comprehend the algorithm method and apply FDO to solve real-world applications quickly. The second issue deals with how to tweak the FDO settings to make the meta-heuristic evolutionary algorithm better in the IoT health service system at evaluating big quantities of information. Ultimately, The target of this chapter's enhancement is to adapt the IoT healthcare framework based on FDO to spawn effective IoT healthcare applications for reasoning out real-world optimization, aggregation, prediction, segmentation, and other technological problems.

Enhancing Algorithm Selection through Comprehensive Performance Evaluation: Statistical Analysis of Stochastic Algorithms

“…However, it is prone to issues such as vulnerability to schedules, occasional entrapment in local optima, and sensitivity to mutation and crossover strategies [34]; I FDO improves individual positions by adding velocity to their current locations, drawing from PSO principles and also influenced by bees' swarming behavior and collaborative decision-making. However, FDO's drawback lies in limited exploration, slow convergence, and sensitivity to proposal distribution [43]; I LPB enhances computational complexity for high school graduates' university transition and study behaviors using genetic algorithm operators. It is versatile and adaptable to different optimization tasks and problem domains, making it a versatile choice.…”