Development of Recommender Systems for Better Services and Products using Data Science DOI

S. Padmapriya,

R Thamizhamuthu,

S. Jagadeesh

et al.

Published: July 6, 2023

Online social networking and e-commerce are becoming increasingly popular. Recommender Systems (RS) let users find relevant information from several possibilities. Internet applications currently need RS. This technology uses huge data to provide customized suggestions improve customer happiness. Concerns ideas help customers choose items. Sentiment Analysis (SA) may increase RS recommendation accuracy by improving user behaviour, views, responses. solves overload in retrieval, but sparsity remains a big problem. SA is notable for reading text expressing preferences. It helps E-Commerce monitor product feedback understand what client wants their research presents hybrid approach correctness. The beats standard models assessment criteria. Modern retailing businesses' operations impossible without RSs. content-based context-aware techniques hybridized providing promising results. Content-based approaches connect consumers new things based on prior ratings activities. Create profiles classify it. Knowledge-based algorithms propose items with minimal use history. These systems case-based recommendations or limitations make recommendations. Finally, ensemble recommender combine source prediction power.

Language: Английский

An applied deep reinforcement learning approach to control active networked microgrids in smart cities with multi-level participation of battery energy storage system and electric vehicles DOI Creative Commons
Reza Sepehrzad, Amir Saman Godazi Langeroudi, Amin Khodadadi

et al.

Sustainable Cities and Society, Journal Year: 2024, Volume and Issue: 107, P. 105352 - 105352

Published: March 18, 2024

This study proposed an intelligent energy management strategy for islanded networked microgrids (NMGs) in smart cities considering the renewable sources uncertainties and power fluctuations. Energy of active frequency control approach is based on probabilistic wavelet fuzzy neural network-deep reinforcement learning algorithm (IPWFNN-DRLA). The formulated with deep Markov decision process solved by soft actor-critic algorithm. NMG local controller (NMGLC) provides information such as frequency, power, generation data, status electric vehicle's battery storage system to central (NMGCC). Then NMGCC calculates support IPWFNN-DRLA sends results NMGLC. model developed a continuous problem-solving space two structures offline training decentralized distributed operation. For this purpose, each has agent (NMGCA) IPWFNN algorithm, NMGCA online back-propagation demonstrates computation accuracy exceeding 98%, along 7.82% reduction computational burden 61.1% time compared alternative methods.

Language: Английский

Citations

27
Fuzzy Q-learning-based approach for real-time energy management of home microgrids using cooperative multi-agent system DOI
Azam Salari, Seyed Ehsan Ahmadi, Mousa Marzband

et al.

Sustainable Cities and Society, Journal Year: 2023, Volume and Issue: 95, P. 104528 - 104528

Published: April 19, 2023

Language: Английский

Citations

32
Risk-involved dominant optimization of multi-energy CCHP-P2G-based microgrids integrated with a variety of storage technologies DOI
Liyuan Zhang, Qiqi Jin, Weichen Zhang

et al.

Journal of Energy Storage, Journal Year: 2024, Volume and Issue: 80, P. 110260 - 110260

Published: Jan. 4, 2024

Language: Английский

Citations

13
Smartgrid-based hybrid digital twins framework for demand side recommendation service provision in distributed power systems DOI

Abiodun E. Onile,

Eduard Petlenkov, Yoash Levron

et al.

Future Generation Computer Systems, Journal Year: 2024, Volume and Issue: 156, P. 142 - 156

Published: March 11, 2024

Language: Английский

Citations

10
A risk-averse day-ahead bidding strategy of transactive energy sharing microgrids with data-driven chance constraints DOI
Yubin Wang, Qiang Yang, Yue Zhou

et al.

Applied Energy, Journal Year: 2023, Volume and Issue: 353, P. 122093 - 122093

Published: Oct. 23, 2023

Language: Английский

Citations

16
Enhancing Energy Systems and Rural Communities through a System of Systems Approach: A Comprehensive Review DOI Creative Commons
Abdellatif Soussi, Enrico Zero, Alessandro Bozzi

et al.

Energies, Journal Year: 2024, Volume and Issue: 17(19), P. 4988 - 4988

Published: Oct. 6, 2024

Today’s increasingly complex energy systems require innovative approaches to integrate and optimize different sources technologies. In this paper, we explore the system of (SoS) approach, which provides a comprehensive framework for improving systems’ interoperability, efficiency, resilience. By examining recent advances in various sectors, including photovoltaic systems, electric vehicles, storage, renewable energy, smart cities, rural communities, study highlights essential role SoSs addressing challenges transition. The principal areas interest include integration advanced control algorithms machine learning techniques development robust communication networks manage interactions between interconnected subsystems. This also identifies significant associated with large-scale SoS implementation, such as real-time data processing, decision-making complexity, need harmonized regulatory frameworks. outlines future directions intelligence autonomy subsystems, are achieving sustainable, resilient, adaptive infrastructure.

Language: Английский

Citations

6
Exploring the role of energy Communities: A comprehensive review DOI Creative Commons
Muhammad Amin, Renato Procopio, Marco Invernizzi

et al.

Energy Conversion and Management X, Journal Year: 2025, Volume and Issue: unknown, P. 100883 - 100883

Published: Jan. 1, 2025

Language: Английский

Citations

0
Leveraging Sustainable Household Energy and Environment Resources Management with Time-Series DOI Creative Commons
José Cecílio, Tiago Rodrigues, M. Barros

et al.

Scientific Data, Journal Year: 2025, Volume and Issue: 12(1)

Published: March 22, 2025

This paper presents a novel and extensive dataset featuring comprehensive cross-sectional data from 13 households with nearly three years of electrical load, energy cost, on-premises solar production directly linked to irradiation weather parameters (SHEERM dataset). The is essential for understanding optimizing utilization achieve Sustainable Development Goals (SDG) 7, 9, 11 13. It provides about production, conditions, residential needs, market prices. combination these variables facilitates multifaceted analysis, fostering advancements in renewable forecasting, climate-sensitive environments, grid management, policy formulation. details the collection process, including sources methodologies employed. Following established literature, we developed implemented machine learning models that comprehensively validate data. Furthermore, as usage notes, offer additional results by applying machine-learning approaches provided aims help design new systems enhance sustainable strategies demonstrate their potential accelerate transition toward carbon neutrality.

Language: Английский

Citations

0
Demand response with incomplete information: A systematic review DOI
Lidong Huang, Hui Liu, Bin Liu

et al.

Electric Power Systems Research, Journal Year: 2025, Volume and Issue: 246, P. 111720 - 111720

Published: April 15, 2025

Language: Английский

Citations

0
Assessing the Feasibility of Integrating Renewable Energy: Decision Tree Analysis for Parameter Evaluation and LSTM Forecasting for Solar and Wind Power Generation in a Campus Microgrid DOI Creative Commons
Fathi Farah Fadoul,

Abdoulaziz Ahmed Hassan,

Ramazan Çağlar

et al.

IEEE Access, Journal Year: 2023, Volume and Issue: 11, P. 124690 - 124708

Published: Jan. 1, 2023

The world has embarked on a road to sustainable energy production. As result, countries have turned microgrid developments. This article aims study the feasibility of renewable sources such as solar PV and wind power for integrating campus, taking example case in East Africa, precisely University Djibouti. We applied weather parameters evaluate potential with Decision Tree method analyzing classifying degrees radiation consistency speed. These data are spread over eight years establish capture seasonal changes prove accessibility specific site. results were compared Random Forest, Logistic Regression, K-Nearest Neighbors, Support Vector Machine, Naïve Bayes classifiers, which showed that performance tree outperformed all other methods an accuracy 0.99321. second work this explored forecasting possible powers predicted LSTM deep learning by generation Solar array turbines simulated PVLib Windpowerlib. favorable, performed well different hyperparameters. With combination machine learning, it was theoretically conclude integration energies since we investigated possibilities evaluating meteorological predictions. Finally, decision scores from architecture features integrated form hybrid Tree-LSTM fusion method. It introduces novel architectural concept can effectively address sequential harness non-linear capabilities trees. proposed model validated tuning Enhancing maximum depth increases at certain point, conversely, reducing minimum sample split improves performance. contributions will help create systems increase transition clean CO2 environment.

Language: Английский

Citations

8