In machine learning, the importance of relevant data increases exponentially. In our proposed approach, we introduce an optimization method that combines Particle Swarm Optimization (PSO) and the firefly algorithm (FA...
详细信息
In machine learning, the importance of relevant data increases exponentially. In our proposed approach, we introduce an optimization method that combines Particle Swarm Optimization (PSO) and the firefly algorithm (FA) to enhance feature selection using decision tree-based classification. PSO is well-suited for small search spaces, while the firefly algorithm is effective for large search spaces. The proposed method, PSOFA-DT aims to improve classification performance by reducing dimensionality and optimizing feature selection. PSO's global search capabilities are complemented by FA's localized search, and the algorithm's effectiveness is evaluated using decision tree accuracy and hold-out cross-validation. Experimental results demonstrate that PSOFA-DT outperforms individual implementations of PSO and FA in feature reduction and classification accuracy. Decision tree accuracy is used as the primary fitness metric, while the firefly algorithm refines the feature selection process. The algorithm balances exploration and exploitation by adjusting key parameters such as inertia weight, learning factors, and attraction coefficients. The firefly algorithm further optimizes feature selection, enhancing decision tree performance.
During the COVID-19 pandemic, effective monitoring and prediction of the evolution of the epidemic are crucial for public health safety. However, traditional monitoring methods exhibit significant shortcomings in data...
详细信息
During the COVID-19 pandemic, effective monitoring and prediction of the evolution of the epidemic are crucial for public health safety. However, traditional monitoring methods exhibit significant shortcomings in data collection efficiency and energy consumption management. In response to this challenge, this research explores the energy consumption issue of mobile edge devices monitoring residents' health data in densely populated urban centers, aiming to minimize the energy consumption and charging time of edge devices while ensuring comprehensive and efficient collection of individual health data. To achieve this goal, this paper proposes a novel optimization firefly algorithm based on the Deep Q-Network (DQN_FA), which is used to optimize the deployment strategy of mobile edge devices. Through simulation experiments, we compare five mobile edge device deployment strategies based on different swarm intelligence optimization algorithms. The experimental results show that the DQN_FA algorithm exhibits comprehensive advantages in performance compared to other intelligent optimization algorithms.
Internet and communication technologies now rely heavily on cloud computing, making it an indispensable component and allowing users to access infrastructure, platforms, and applications on a flexible, scalable, on-de...
详细信息
Agricultural workers, during farming, have to perform many activities in hot summer, heavy rain, and in cold winter. The activities performed require a high physical workload. Farming activities are repetitive in natu...
详细信息
In this research, firefly algorithm to generate summary of a given video is proposed. The firefly algorithm is an optimization algorithm that draws inspiration from the behaviour of fireflies. There are many types of ...
详细信息
Network intrusion detection is the process of identifying malicious activity in a network by analyzing the network traffic behavior. Data mining techniques are widely used in Intrusion Detection System (IDS) to detect...
详细信息
Network intrusion detection is the process of identifying malicious activity in a network by analyzing the network traffic behavior. Data mining techniques are widely used in Intrusion Detection System (IDS) to detect anomalies. Dimensionality reduction plays a vital role in IDS, since detecting anomalies from high dimensional network traffic feature is time-consuming process. Feature selection influences the speed of the analysis and the proposed work, deploys filter and wrapper based method with firefly algorithm in the wrapper for selecting the features. The resulting features are subjected to C4.5 and Bayesian Networks (BN) based classifier with KDD CUP 99 dataset. The experimental results show that 10 features are sufficient to detect the intrusion showing improved accuracy. The proposed work is compared with the existing work showing promising improvements. (C) 2018 Elsevier Ltd. All rights reserved.
Various topologies of the Direct Current (DC) microgrid have been proposed in the literature by considering practical requirements. One of the important criteria in topology selection is the availability of a well-fun...
详细信息
Various topologies of the Direct Current (DC) microgrid have been proposed in the literature by considering practical requirements. One of the important criteria in topology selection is the availability of a well-functioning protection scheme. A significant part of the DC microgrid protection system literature considers differential protection schemes for various topologies. However this scheme doesn't work satisfactorily with noisy measurements. Noisy measurements are a reality and can not be avoided for any real engineering system. This article proposes a robust protection scheme to alleviate some problems associated with topology selection. The proposed scheme considers noisy current measurements and works by estimating the time derivative of line currents. The proposed approach uses continuous finite-time convergent differentiator to estimate the derivative. The gains of the differentiator were tuned through firefly algorithm-based optimization technique. Matlab/Simulink (R)-based simulations verified the effectiveness of the proposed approach over the existing differential current scheme.
A recently developed metaheuristic optimization algorithm, firefly algorithm (FA), mimics the social behavior of fireflies based on the flashing and attraction characteristics of fireflies. In the present study, we wi...
详细信息
A recently developed metaheuristic optimization algorithm, firefly algorithm (FA), mimics the social behavior of fireflies based on the flashing and attraction characteristics of fireflies. In the present study, we will introduce chaos into FA so as to increase its global search mobility for robust global optimization. Detailed studies are carried out on benchmark problems with different chaotic maps. Here, 12 different chaotic maps are utilized to tune the attractive movement of the fireflies in the algorithm. The results show that some chaotic FAs can clearly outperform the standard FA. (c) 2012 Elsevier B.V. All rights reserved.
In this paper, firefly algorithm (FA) for optimal tuning of PI controllers for load frequency control of hybrid system composing of photovoltaic (PV) system and thermal generator is introduced. Also, maximum power poi...
详细信息
In this paper, firefly algorithm (FA) for optimal tuning of PI controllers for load frequency control of hybrid system composing of photovoltaic (PV) system and thermal generator is introduced. Also, maximum power point tracking of PV is considered in the design process. The block diagram of the hybrid system is performed. To robustly tune the parameters of controllers, a time domain-based objective function is established which is solved by the FA. Simulation results are presented to show the improved performance of the suggested FA-based controllers compared with genetic algorithm (GA). These results show that the proposed controllers present better performance over GA in terms of settling times and different indices.
A particle filter (PF) has been considered one of the most useful tools for nonlinear non-Gaussian systems. However, the estimation accuracy is limited by sample impoverishment due to resampling. Therefore, a firefly ...
详细信息
A particle filter (PF) has been considered one of the most useful tools for nonlinear non-Gaussian systems. However, the estimation accuracy is limited by sample impoverishment due to resampling. Therefore, a firefly algorithm-based PF is proposed to solve this problem. In the proposed algorithm, the resampling step is performed based on the firefly algorithm. Finally, simulations are conducted to illustrate the superior performance of the proposed algorithm over that of a PF and a regularized particle filter.
暂无评论