The impact of ChatGPT on higher education

Frontiers in Education, Journal Year: 2023, Volume and Issue: 8

Published: Sept. 8, 2023

Introduction This study explores the effects of Artificial Intelligence (AI) chatbots, with a particular focus on OpenAI’s ChatGPT, Higher Education Institutions (HEIs). With rapid advancement AI, understanding its implications in educational sector becomes paramount. Methods Utilizing databases like PubMed, IEEE Xplore, and Google Scholar, we systematically searched for literature AI chatbots’ impact HEIs. Our criteria prioritized peer-reviewed articles, prominent media outlets, English publications, excluding tangential chatbot mentions. After selection, data extraction focused authors, design, primary findings. The analysis combined descriptive thematic approaches, emphasizing patterns applications chatbots Results review revealed diverse perspectives ChatGPT’s potential education. Notable benefits include research support, automated grading, enhanced human-computer interaction. However, concerns such as online testing security, plagiarism, broader societal economic impacts job displacement, digital literacy gap, AI-induced anxiety were identified. also underscored transformative architecture ChatGPT versatile sector. Furthermore, advantages streamlined enrollment, improved student services, teaching enhancements, aid, increased retention highlighted. Conversely, risks privacy breaches, misuse, bias, misinformation, decreased human interaction, accessibility issues Discussion While AI’s global expansion is undeniable, there pressing need balanced regulation application within Faculty members are encouraged to utilize tools proactively ethically mitigate risks, especially academic fraud. Despite study’s limitations, including an incomplete representation overall effect education absence concrete integration guidelines, it evident that technologies present both significant risks. advocates thoughtful responsible

Language: Английский

---

Prediction of Photovoltaic Power by the Informer Model Based on Convolutional Neural Network

Sustainability, Journal Year: 2022, Volume and Issue: 14(20), P. 13022 - 13022

Published: Oct. 12, 2022

Accurate prediction of photovoltaic power is great significance to the safe operation grids. In order improve accuracy, a similar day clustering convolutional neural network (CNN)–informer model was proposed predict power. Based on correlation analysis, it determined that global horizontal radiation meteorological factor had greatest impact power, and dataset divided into four categories according between factors fluctuation characteristics; then, CNN used extract feature information trends different subsets, features output by were fused input informer model. The establish temporal relationship historical data, final generation result obtained. experimental results show CNN–informer method has high accuracy stability in outperforms other deep learning methods.

Language: Английский

---

Multi-energy load forecasting for integrated energy system based on sequence decomposition fusion and factors correlation analysis

Energy, Journal Year: 2024, Volume and Issue: 308, P. 132796 - 132796

Published: Aug. 14, 2024

Language: Английский

---

Few-Shot Load Forecasting Under Data Scarcity in Smart Grids: A Meta-Learning Approach

Energies, Journal Year: 2025, Volume and Issue: 18(3), P. 742 - 742

Published: Feb. 6, 2025

Despite the rapid expansion of smart grids and large volumes data at individual consumer level, there are still various cases where adequate collection to train accurate load forecasting models is challenging or even impossible. This paper proposes adapting an established Model-Agnostic Meta-Learning algorithm for short-term in context few-shot learning. Specifically, proposed method can rapidly adapt generalize within any unknown time series arbitrary length using only minimal training samples. In this context, meta-learning model learns optimal set initial parameters a base-level learner recurrent neural network. The evaluated dataset historical consumption from real-world consumers. examined series’ short length, it produces forecasts outperforming transfer learning task-specific machine methods by 12.5%. To enhance robustness fairness during evaluation, novel metric, mean average log percentage error, that alleviates bias introduced commonly used MAPE metric. Finally, studies evaluate model’s under different hyperparameters lengths also conducted, demonstrating approach consistently outperforms all other models.

Language: Английский

---

A New Deep Learning Restricted Boltzmann Machine for Energy Consumption Forecasting

Sustainability, Journal Year: 2022, Volume and Issue: 14(16), P. 10081 - 10081

Published: Aug. 15, 2022

A key issue in the desired operation and development of power networks is knowledge load growth electricity demand coming years. Mid-term forecasting (MTLF) has an important rule planning optimal use systems. However, MTLF a complicated problem, lot uncertain factors variables disturb consumption pattern. This paper presents practical approach for MTLF. new deep learning restricted Boltzmann machine (RBM) proposed modelling energy consumption. The contrastive divergence algorithm presented tuning parameters. All parameters RBMs, number input variables, type inputs, also layer neuron numbers are optimized. statistical suggested to determine effective variables. In addition climate such as temperature humidity, effects other economic investigated. Finally, using simulated real-world data examples, it shown that one year ahead, mean absolute percentage error (MAPE) peak less than 5%. Moreover, 24-h pattern forecasting, MAPE all days

Language: Английский

---

Seq2Seq-LSTM With Attention for Electricity Load Forecasting in Brazil

IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 30020 - 30029

Published: Jan. 1, 2024

Electricity load forecasting is important to planning the decision-making regarding use of energy resources, in which power system must be operated guarantee supply electricity future at lowest possible price. With rise ability based on deep learning, these approaches are promising this context. Considering attention mechanism capture long-range dependencies, it highly recommended for sequential data processing, where time series-related tasks stand out. a sequence-to-sequence (Seq2Seq) series Brazil, paper proposes long short-term memory (LSTM) with perform forecasting. The proposed Seq2Seq-LSTM outperforms other well-established models. Having mean absolute error equal 0.3027 method shown field applications. contributes by implementing an Seq2Seq data, therefore, more than one correlated signal can used prediction enhancing its capacity when avaliable.

Language: Английский

---

Resilient Electricity Load Forecasting Network with Collective Intelligence Predictor for Smart Cities

Electronics, Journal Year: 2024, Volume and Issue: 13(4), P. 718 - 718

Published: Feb. 9, 2024

Accurate electricity forecasting is essential for smart cities to maintain grid stability by allocating resources in advance, ensuring better integration with renewable energies, and lowering operation costs. However, most models that use machine learning cannot handle the missing values possess a single point of failure. With rapid technological advancement, are becoming lucrative targets cyberattacks induce packet loss or take down servers offline via distributed denial-of-service attacks, disrupting system inducing load data. This paper proposes collective intelligence predictor, which uses modular three-level networks decentralize strengthen against values. Compared existing models, it achieves coefficient determination score 0.98831 no using base model Level 0 network. As forecasted zone rise 90% single-model method longer effective, 0.89345 meta-model 1 network aggregate results from 0. Finally, as reach 100%, 0.81445 reconstructing forecast other zones 2

Language: Английский

---

Artificial intelligence-based forecasting models for integrated energy system management planning: An exploration of the prospects for South Africa

Energy Conversion and Management X, Journal Year: 2024, Volume and Issue: 24, P. 100772 - 100772

Published: Oct. 1, 2024

Language: Английский

---

Short-Term Electricity Load Forecasting Using the Temporal Fusion Transformer: Effect of Grid Hierarchies and Data Sources

Published: June 16, 2023

Recent developments related to the energy transition pose particular challenges for distribution grids. Hence, precise load forecasts become more and important effective grid management. Novel modeling approaches such as Transformer architecture, in Temporal Fusion (TFT), have emerged promising methods time series forecasting. To date, just a handful of studies apply TFTs electricity forecasting problems, mostly considering only single datasets few covariates. Therefore, we examine potential TFT architecture hourly short-term across different horizons (day-ahead week-ahead) network levels (grid substation level). We find that does not offer higher predictive performance than state-of-the-art LSTM model day-ahead on entire grid. However, results display significant improvements when applied at level with subsequent aggregation upper grid-level, resulting prediction error 2.43% (MAPE) best-performing scenario. In addition, appears remarkable over approach week-ahead (yielding 2.52% lowest). outline avenues future research using forecasting, including exploration various (e.g., grid, substation, household

Language: Английский

---

Deep learning model-transformer based wind power forecasting approach

Frontiers in Energy Research, Journal Year: 2023, Volume and Issue: 10

Published: Jan. 16, 2023

The uncertainty and fluctuation are the major challenges casted by large penetration of wind power (WP). As one most important solutions for tackling these issues, accurate forecasting is able to enhance energy consumption improve rate WP. In this paper, we propose a deep learning model-transformer based (WPF) model. transformer neural network architecture on attention mechanism, which clearly different from other models such as CNN or RNN. basic unit consists residual structure, self-attention mechanism feedforward network. overall multilayer encoder decoder structure enables complete modeling sequential data. By comparing results with four models, LSTM, accuracy efficiency have been validated. Furthermore, migration experiments show that can also provide good performance.

Language: Английский

---