Electric Vehicle Charging Station Recommendations Considering User Charging Preferences Based on Comment Data DOI Creative Commons

Houzhi Li,

Qingwen Han,

Xueyuan Bai

et al.

Energies, Journal Year: 2024, Volume and Issue: 17(21), P. 5514 - 5514

Published: Nov. 4, 2024

User preferences are important for electric vehicle charging station (EVCS) recommendations, but they have not been deeply analyzed. Therefore, in this study, user identified and applied to EVCS recommendations using a hybrid model that integrates LightGBM singular value decomposition (SVD). In the model, is used predict ratings according users’ comments regarding orders, feature importance reported by each output. Then, co-occurrence matrix between users stations (EVCSs) constructed decomposed SVD. Based on results, final evaluated scores of EVCSs can be calculated. Upon ranking scores, recommendation results obtained, taking into account preferences. The sample data consist 28,306 orders from 508 at 241 Linyi, Shandong, China. experimental show proposed outperforms benchmark models terms precision, recall, F1 score, its score increased 96% compared with traditional item-based collaborative filtering method counts recommendations.

Language: Английский

Developing an IoT-driven delta robot to stimulate the growth of mulberry branch cuttings cultivated aeroponically using machine vision technology DOI
Osama Elsherbiny, Jianmin Gao, Ming Ma

et al.

Computers and Electronics in Agriculture, Journal Year: 2025, Volume and Issue: 232, P. 110111 - 110111

Published: Feb. 11, 2025

Language: Английский

Citations

1
Global Potato Production Forecasting Based on Time Series Analysis and Advanced Waterwheel Plant Optimization Algorithm DOI
Amel Ali Alhussan, Doaa Sami Khafaga, Mostafa Abotaleb

et al.

Potato Research, Journal Year: 2024, Volume and Issue: 67(4), P. 1965 - 2000

Published: May 8, 2024

Language: Английский

Citations

4
Real-Time Electric Taxi Guidance for Battery Swapping Stations Under Dynamic Demand DOI Creative Commons
Feng Yu, Xiaochun Lu, Xiaohui Huang

et al.

Energies, Journal Year: 2025, Volume and Issue: 18(9), P. 2193 - 2193

Published: April 25, 2025

High battery swapping demand from electric taxis and drivers’ subjective station selection often leads to congestion the uneven utilization of stations (BSSs). Efficient vehicle guidance is essential for improving operational performance taxis. In this study, we have developed a vehicle-to-station model that considers dynamic diverse driver response-time preferences. We proposed two decision-making strategies BSS recommendations. The first real-time optimization method uses greedy algorithm provide immediate guidance. second delayed framework performs batch scheduling under high demand. It integrates genetic with KD-tree search handle insertion. A case study based on Beijing’s Fourth Ring Road network was conducted evaluate four preference scenarios. results show clear differences in waiting times. balanced consideration travel distance, time, cost can effectively reduce delays drivers improve utilization. This research provides practical approach systems.

Language: Английский

Citations

0
Exploring the role of artificial intelligence in enhancing battery performance and mitigating cybersecurity threats in electric vehicles: A systematic literature review DOI Open Access
Husni Abdillah,

N.A.H. Wildan Rizkia,

Sidharta Sidharta

et al.

Procedia Computer Science, Journal Year: 2024, Volume and Issue: 245, P. 155 - 165

Published: Jan. 1, 2024

Language: Английский

Citations

2
Application of graph modeling and contrast learning in recommender system DOI Creative Commons
Wentao Zhang

Applied and Computational Engineering, Journal Year: 2024, Volume and Issue: 64(1), P. 50 - 55

Published: May 14, 2024

With the wide application of personalized recommender system in various fields, how to improve accuracy and level has become a research hotspot. In this paper, method combining graph modeling contrast learning is proposed performance recommendation by mining complex user project interaction preference. We first construct user-project graph, extract features structure neural network (GNN) . particular, convolution (GCN) used update node representation, comparative introduced optimize feature representation so as personalization recommendation. The experimental results show that superior traditional accuracy, recall F 1 score. By analyzing mechanism learning, paper further expounds theoretical basis practical improving system, points out limitations existing methods future direction.

Language: Английский

Citations

0
Urban Electric Vehicle Charging Station Placement Optimization with Graylag Goose Optimization Voting Classifier DOI Open Access
Amel Ali Alhussan, Doaa Sami Khafaga,

El-Sayed M. El-kenawy

et al.

Computers, materials & continua/Computers, materials & continua (Print), Journal Year: 2024, Volume and Issue: 80(1), P. 1163 - 1177

Published: Jan. 1, 2024

To reduce the negative effects that conventional modes of transportation have on environment, researchers are working to increase use electric vehicles. The demand for environmentally friendly may be hampered by obstacles such as a restricted range and extended rates recharge. establishment urban charging infrastructure includes both fast ultra-fast terminals is essential address this issue. Nevertheless, powering these presents challenges because high energy requirements, which influence quality service. Modelling maximum hourly capacity each station based its geographic location necessary arrive at an accurate estimation resources required infrastructure. It vital do analysis specific regional traffic patterns, road networks, route details, junction density, economic zones, rather than making arbitrary conclusions about patterns. When vehicle simulated using data other variables, it possible detect limits in design current engineering system. Initially, binary graylag goose optimization (bGGO) algorithm utilized purpose feature selection. Subsequently, (GGO) voting classifier decision allocate stations while taking into consideration cost variable congestion. Based results variance (ANOVA), comprehensive summary components contribute observed variability dataset provided. Wilcoxon Signed Rank Test compare actual median accuracy values several different algorithms, GGO algorithm, grey wolf (GWO), whale (WOA), particle swarm (PSO), firefly (FA), genetic (GA), theoretical would expected there no difference.

Language: Английский

Citations

0
Enhancing Student Performance Prediction with Greylag Goose Optimization Algorithm DOI

Faris H. Rizk,

Mahmoud Elshabrawy Mohamed,

Basant Sameh

et al.

2022 International Telecommunications Conference (ITC-Egypt), Journal Year: 2024, Volume and Issue: unknown, P. 32 - 37

Published: July 22, 2024

Language: Английский

Citations

0
Recommender systems in smart campus: a systematic mapping DOI
Martin Hideki Mensch Maruyama, Luan Silveira, Elvandi da Silva Júnior

et al.

Knowledge and Information Systems, Journal Year: 2024, Volume and Issue: unknown

Published: Sept. 26, 2024

Language: Английский

Citations

0
Regulating thermal patient suit in operation room by artificial intelligence technique DOI

Hassan Flaeh Rdhaiwi,

Ahmed R. Ajel,

Saleem Lateef Mohammed

et al.

AIP conference proceedings, Journal Year: 2024, Volume and Issue: 3232, P. 040007 - 040007

Published: Jan. 1, 2024

Language: Английский

Citations

0
Electric Vehicle Charging Station Recommendations Considering User Charging Preferences Based on Comment Data DOI Creative Commons

Houzhi Li,

Qingwen Han,

Xueyuan Bai

et al.

Energies, Journal Year: 2024, Volume and Issue: 17(21), P. 5514 - 5514

Published: Nov. 4, 2024

User preferences are important for electric vehicle charging station (EVCS) recommendations, but they have not been deeply analyzed. Therefore, in this study, user identified and applied to EVCS recommendations using a hybrid model that integrates LightGBM singular value decomposition (SVD). In the model, is used predict ratings according users’ comments regarding orders, feature importance reported by each output. Then, co-occurrence matrix between users stations (EVCSs) constructed decomposed SVD. Based on results, final evaluated scores of EVCSs can be calculated. Upon ranking scores, recommendation results obtained, taking into account preferences. The sample data consist 28,306 orders from 508 at 241 Linyi, Shandong, China. experimental show proposed outperforms benchmark models terms precision, recall, F1 score, its score increased 96% compared with traditional item-based collaborative filtering method counts recommendations.

Language: Английский

Citations

0