Towards a semi-local random walk technique through multilayer social networks to improve link prediction

Journal of Complex Networks, Journal Year: 2023, Volume and Issue: 12(1)

Published: Dec. 22, 2023

Abstract The rapid expansion of social networks has generated a growing need for scalable algorithms capable effectively predicting links. Link prediction is crucial area study within complex research. aims to predict future connections between nodes from the current snapshot network and plays vital role in estimating growth networks. This article introduces an improved approach link by exploiting extended version local random walk as semi-local (SLRW) multilayer Here, taking into account connectivity structural similarity involved nodes, we propose SLRW method acquire sequence with highest similarity. Also, metric includes distributed technique identify nearest neighbours considering neighbourhood concept. To ensure optimal performance, conduct extensive studies on various hyperparameters proposed metric. experimental results conducted different datasets demonstrate that achieves improvements field compared state-of-the-art baselines.

Language: Английский

---

An efficient dynamic service provisioning mechanism in fog computing environment: A learning automata approach

Expert Systems with Applications, Journal Year: 2022, Volume and Issue: 198, P. 116863 - 116863

Published: March 18, 2022

Language: Английский

---

An Efficient Policy-Based Scheduling and Allocation of Virtual Machines in Cloud Computing Environment

Journal of Electrical and Computer Engineering, Journal Year: 2022, Volume and Issue: 2022, P. 1 - 12

Published: Sept. 24, 2022

Cloud computing has become the most challenging research field in current information technology scenario. In this, a set of user tasks are scheduled and allocated to numerous kinds heterogeneous virtual machines (VMs) cloud data centers (CDCs), these VMs hosted by diverse types physical (PMs). It extends several features, encompassing elasticity, safety, sustainability, even adequate maintenance compared traditional centers. There techniques available for VM scheduling allocation. However, it still requires existence new mechanisms that can reduce execution time (ET) tasks, improve optimization energy usage resource utilization (RU), consumption. Along with optimization, (VMS) allocation (VMA) two-level issues need be considered essential policies govern mechanisms. Hence, executing optimal VMS VMA center, methodologies, such as enhanced shark smell algorithm (ESSOA) at first level Brownian movement-centered gravitation search (BMGSA) second level, proposed this work define policies. Firstly, requests reserved on appropriate PM ESSOA, which lowest cost within deadline limits, BMGSA decides chosen limitations level. To demonstrate algorithm’s efficiency, simulations carried out using Java language-based CloudSim simulator, mechanism outcomes acquired existing approaches. The simulation results show suggested is efficient terms cost, degree imbalance (DOI), make span (MS), (RU).

Language: Английский

---

Energy-efficient virtual machine placement in distributed cloud using NSGA-III algorithm

Journal of Cloud Computing Advances Systems and Applications, Journal Year: 2023, Volume and Issue: 12(1)

Published: Aug. 26, 2023

Abstract Cloud computing is the most widely adapted model to process scientific workloads in remote servers accessed through internet. In IaaS cloud, virtual machine (VM) execution unit that processes user workloads. Virtualization enables of multiple machines (VMs) on a single physical (PM). Virtual placement (VMP) strategically assigns VMs suitable devices within data center. From cloud provider's perspective, must be placed optimally reduce resource wastage aid economic revenue and develop green centres. providers need an efficient methodology minimize wastage, power consumption, network transmission delay. This paper uses NSGA-III, multi-objective evolutionary algorithm, simultaneously mentioned objectives obtain non-dominated solution. The performance metrics (Overall Nondominated Vector Generation Spacing) proposed NSGA-III algorithm compared with other algorithms, namely VEGA, MOGA, SPEA, NSGA-II. It observed performs 7% better existing terms ONVG 12% results spacing. ANOVA DMRT statistical tests are used cross-validate results.

Language: Английский

---

Mixed-type wafer defect detection based on multi-branch feature enhanced residual module

Expert Systems with Applications, Journal Year: 2023, Volume and Issue: 242, P. 122795 - 122795

Published: Nov. 29, 2023

Language: Английский

---