Combination of PSO and RF with Weights for Supervised Learning Model DOI
Wei‐Chang Yeh, Chia‐Ling Huang

Published: Oct. 12, 2023

The pattern, path, and discrimination according to the data can automatically discovered by mathematical models of machine learning (ML), accordingly outcomes are applied project prospects and/or cause decisions as stated brand-fresh, unseen data. supervised (SL) makes whole solutions identified while generating projects about solution gathering information based on labeled most popular SL method is random forests (RF) that adaptable may be used solve both grouping regression issues. RF training procedure lengthy, resource centralized, prone wrong group a result these other drawbacks. In this context, combination particle swarm optimization (PSO) weighted presented in order improve efficacy RF.

Language: Английский

Independent task scheduling algorithms in fog environments from users’ and service providers’ perspectives: a systematic review DOI
Abdulrahman K. Al-Qadhi, Rohaya Latip, Raymond Chiong

et al.

Cluster Computing, Journal Year: 2025, Volume and Issue: 28(3)

Published: Jan. 28, 2025

Language: Английский

Citations

1
PGA: A New Hybrid PSO and GA Method for Task Scheduling with Deadline Constraints in Distributed Computing DOI Creative Commons

Kaili Shao,

Ying Song, Bo Wang

et al.

Mathematics, Journal Year: 2023, Volume and Issue: 11(6), P. 1548 - 1548

Published: March 22, 2023

Distributed computing, e.g., cluster and cloud has been applied in almost all areas for data processing, while high resource efficiency user satisfaction are still the ambition of distributed computing. Task scheduling is indispensable achieving goal. As task problem NP-hard, heuristics meta-heuristics frequently applied. Every method its own advantages limitations. Thus, this paper, we designed a hybrid heuristic by exploiting global search ability Genetic Algorithm (GA) fast convergence Particle Swarm Optimization (PSO). Different from existing approaches that simply sequentially perform two or more algorithms, PGA applies evolutionary GA integrates self- social cognitions into evolution. We conduct extensive simulated environments performance evaluation, where simulation parameters set referring to some recent related works. Experimental results show 27.9–65.4% 33.8–69.6% better than several works, on average, efficiency, respectively.

Language: Английский

Citations

18
High-Accuracy Analytical Model for Heterogeneous Cloud Systems with Limited Availability of Physical Machine Resources Based on Markov Chain DOI Open Access
Sławomir Hanczewski, Maciej Stasiak, Michał Weissenberg

et al.

Electronics, Journal Year: 2024, Volume and Issue: 13(11), P. 2161 - 2161

Published: June 1, 2024

The article presents the results of a study on modeling cloud systems. In this research, authors developed both analytical and simulation models. System analysis was conducted at level virtual machine support, corresponding to Infrastructure as Service (IaaS). models assumed that machines different sizes are offered part IaaS, reflecting heterogeneous nature modern Additionally, it due limitations in access physical server resources, only portion these resources could be used create machines. model is based Markov chain for state-dependent system divided into an external structure, represented by collection machines, internal single machine. novel approach determine equivalent traffic, approximating real traffic appearing input under assumptions request distribution. As result, possible actual loss probability entire system. obtained from (simulation analytical) were summarized common graphs. studies related parameters commercially research confirmed high accuracy its independence number instances Thus, can dimension

Language: Английский

Citations

1
Time-reliability optimization for the stochastic traveling salesman problem DOI
Wei‐Chang Yeh

Reliability Engineering & System Safety, Journal Year: 2024, Volume and Issue: 248, P. 110179 - 110179

Published: May 6, 2024

Language: Английский

Citations

0
All You Need to Know About Cloud Elasticity Technologies DOI

Summit Shrestha,

Zheng Song, Yazhi Liu

et al.

Published: Jan. 1, 2023

After more than a decade since the inception of cloud computing, underlying technologies supporting it have experienced significant advancements and now matured enough to provide satisfactory QoS for its users. Among these technologies, particular attention has been given development elasticity, which is prominent feature computing. However, most recent comprehensive survey on elasticity was published in 2017 fails encompass latest progress field. Additionally, there lack understanding regarding interplay different technologies. These create knowledge gap between high-level concept state-of-the-art technical details relevant computing users, developers, researchers. To address this gap, we carefully select 145 influential papers, both classical recent, elasticity. We taxonomy categorize enabling reported papers. For each technology, thoroughly examine limitations. This paper serves as valuable resource researchers practitioners, providing them with review up-to-date research It also provides good foundation enable new practitioners enter field gain an insight into

Language: Английский

Citations

0
Combination of PSO and RF with Weights for Supervised Learning Model DOI
Wei‐Chang Yeh, Chia‐Ling Huang

Published: Oct. 12, 2023

The pattern, path, and discrimination according to the data can automatically discovered by mathematical models of machine learning (ML), accordingly outcomes are applied project prospects and/or cause decisions as stated brand-fresh, unseen data. supervised (SL) makes whole solutions identified while generating projects about solution gathering information based on labeled most popular SL method is random forests (RF) that adaptable may be used solve both grouping regression issues. RF training procedure lengthy, resource centralized, prone wrong group a result these other drawbacks. In this context, combination particle swarm optimization (PSO) weighted presented in order improve efficacy RF.

Language: Английский

Citations

0