Exploding the myths: An introduction to artificial neural networks for prediction and forecasting

Environmental Modelling & Software, Journal Year: 2023, Volume and Issue: 167, P. 105776 - 105776

Published: July 5, 2023

Artificial Neural Networks (ANNs), sometimes also called models for deep learning, are used extensively the prediction of a range environmental variables. While potential ANNs is unquestioned, they surrounded by an air mystery and intrigue, leading to lack understanding their inner workings. This has led perpetuation number myths, resulting in misconception that applying primarily involves "throwing" large amount data at "black-box" software packages. this convenient way side-step principles applied development other types models, comes significant cost terms usefulness models. To address these issues, inroductory overview paper explodes common myths surrounding use outlines state-of-the-art approaches developing enable them be with confidence practice.

Language: Английский

---

Interpretability Research of Deep Learning: A Literature Survey

Information Fusion, Journal Year: 2024, Volume and Issue: 115, P. 102721 - 102721

Published: Oct. 9, 2024

Language: Английский

---

A compound approach for ten-day runoff prediction by coupling wavelet denoising, attention mechanism, and LSTM based on GPU parallel acceleration technology

Earth Science Informatics, Journal Year: 2024, Volume and Issue: 17(2), P. 1281 - 1299

Published: Jan. 10, 2024

Language: Английский

---

A Comprehensive Review on Deep Learning Applications in Advancing Biodiesel Feedstock Selection and Production Processes

Green Energy and Intelligent Transportation, Journal Year: 2025, Volume and Issue: unknown, P. 100260 - 100260

Published: Jan. 1, 2025

Language: Английский

---

A step forward in food science, technology and industry using artificial intelligence

Trends in Food Science & Technology, Journal Year: 2023, Volume and Issue: 143, P. 104286 - 104286

Published: Dec. 4, 2023

Language: Английский

---

Deep Learning for Multi-Source Data-Driven Crop Yield Prediction in Northeast China

Agriculture, Journal Year: 2024, Volume and Issue: 14(6), P. 794 - 794

Published: May 22, 2024

The accurate prediction of crop yields is crucial for enhancing agricultural efficiency and ensuring food security. This study assesses the performance CNN-LSTM-Attention model in predicting maize, rice, soybeans Northeast China compares its effectiveness with traditional models such as RF, XGBoost, CNN. Utilizing multi-source data from 2014 to 2020, which include vegetation indices, environmental variables, photosynthetically active parameters, our research examines model’s capacity capture essential spatial temporal variations. integrates Convolutional Neural Networks, Long Short-Term Memory, an attention mechanism effectively process complex datasets manage non-linear relationships within data. Notably, explores potential using kNDVI multiple crops, highlighting effectiveness. Our findings demonstrate that advanced deep-learning significantly enhance yield accuracy over methods. We advocate incorporation sophisticated technologies practices, can substantially improve production strategies.

Language: Английский

---

A proposed framework for crop yield prediction using hybrid feature selection approach and optimized machine learning

Neural Computing and Applications, Journal Year: 2024, Volume and Issue: 36(33), P. 20723 - 20750

Published: Aug. 16, 2024

Abstract Accurately predicting crop yield is essential for optimizing agricultural practices and ensuring food security. However, existing approaches often struggle to capture the complex interactions between various environmental factors growth, leading suboptimal predictions. Consequently, identifying most important feature vital when leveraging Support Vector Regressor (SVR) prediction. In addition, manual tuning of SVR hyperparameters may not always offer high accuracy. this paper, we introduce a novel framework yields that address these challenges. Our integrates new hybrid selection approach with an optimized model enhance prediction accuracy efficiently. The proposed comprises three phases: preprocessing, selection, phases. preprocessing phase, data normalization conducted, followed by application K-means clustering in conjunction correlation-based filter (CFS) generate reduced dataset. Subsequently, FMIG-RFE proposed. Finally, phase introduces improved variant Crayfish Optimization Algorithm (COA), named ICOA, which utilized optimize thereby achieving superior along approach. Several experiments are conducted assess evaluate performance framework. results demonstrated over state-of-art approaches. Furthermore, experimental findings regarding ICOA optimization algorithm affirm its efficacy model, enhancing both computational efficiency, surpassing algorithms.

Language: Английский

---

A Novel Approach for State of Health Estimation and Remaining Useful Life Prediction of Supercapacitors Using an Improved Honey Badger Algorithm Assisted Hybrid Neural Network

Protection and Control of Modern Power Systems, Journal Year: 2024, Volume and Issue: 9(6), P. 1 - 18

Published: Nov. 1, 2024

Supercapacitors (SCs) are widely recognized as excellent clean energy storage devices. Accurate state of health (SOH) estimation and remaining useful life (RUL) prediction essential for ensuring their safe reliable operation. This paper introduces a novel method SOH RUL prediction, based on hybrid neural network optimized by an improved honey badger algorithm (HBA). The combines the advantages convolutional (CNN) bidirectional long-short-term memory (BiLSTM) network. HBA optimizes hyperparameters CNN automatically extracts deep features from time series data reduces dimensionality, which then used input BiLSTM. Additionally, recurrent dropout is introduced in layer to reduce overfitting facilitate learning process. approach not only improves accuracy estimates forecasts but also significantly processing time. SCs under different working conditions validate proposed method. results show that model effectively features, enriches local details, enhances global perception capabilities. outperforms single models, reducing root mean square error below 1%, offers higher robustness compared other methods.

Language: Английский

---

Unlocking the black box: an in-depth review on interpretability, explainability, and reliability in deep learning

Neural Computing and Applications, Journal Year: 2024, Volume and Issue: unknown

Published: Nov. 18, 2024

Language: Английский

---

Destructive and non-destructive measurement approaches and the application of AI models in precision agriculture: a review

Precision Agriculture, Journal Year: 2024, Volume and Issue: 25(3), P. 1127 - 1180

Published: Feb. 27, 2024

Abstract The estimation of pre-harvest fruit quality and maturity is essential for growers to determine the harvest timing, storage requirements profitability crop yield. In-field indicators are highly variable require high spatiotemporal resolution data, which can be obtained from contemporary precision agriculture systems. Such systems exploit various state-of-the-art sensors, increasingly relying on spectrometry imaging techniques in association with advanced Artificial Intelligence (AI) and, particular, Machine Learning (ML) algorithms. This article presents a critical review estimation, focus destructive non-destructive measurement approaches, applications ML domain. A analysis advantages disadvantages different conducted by surveying recent articles methods discern trends performance applicability. Advanced data-fusion combining information multiple sensors being used develop more accurate representations entire field. achieved incorporating AI algorithms, such as support vector machines, k-nearest neighbour, neural networks, clustering. Based an extensive survey recently published research, also identifies most effective indices, namely: sugar content, acidity firmness. concludes highlighting outstanding technical challenges promising areas future research. Hence, this research has potential provide valuable resource growers, allowing them familiarize themselves Smart Agricultural methodologies currently use. These practices gradually incorporated their perspective, taking into account availability use efficient indices.

Language: Английский

---