Machine learning-based anomaly detection and prediction in commercial aircraft using autonomous surveillance data

PLoS ONE, Journal Year: 2025, Volume and Issue: 20(2), P. e0317914 - e0317914

Published: Feb. 6, 2025

Regarding the transportation of people, commodities, and other items, aeroplanes are an essential need for society. Despite generally low danger associated with various modes transportation, some accidents may occur. The creation a machine learning model employing data from autonomous-reliant surveillance transmissions is detection prediction commercial aircraft accidents. This research included development abnormal categorisation models, assessment recognition quality, anomalies. methodology consisted following steps: formulation problem, selection labelling, construction prediction, installation, testing. tagging technique was based on requirements set by Global Aviation Organisation business jet-engine aircraft, which expert pilots then validated. 93% precision demonstrated excellent match most effective model, linear dipole Furthermore, "good fit" verified its achieved area-under-the-curve ratios 0.97 identification 0.96 daily detection.

Language: Английский

---

Natural Gas Futures Price Prediction Based on Variational Mode Decomposition–Gated Recurrent Unit/Autoencoder/Multilayer Perceptron–Random Forest Hybrid Model

Sustainability, Journal Year: 2025, Volume and Issue: 17(6), P. 2492 - 2492

Published: March 12, 2025

Forecasting natural gas futures prices can help to promote sustainable global energy development, as the efficient use of a clean source has become key growing demand for development. This study proposes new hybrid model prediction prices. Firstly, original price series is decomposed, and subsequences, along with influencing factors, are used input variables. Secondly, variables grouped based on their correlations output variable, different models employed forecast each group. A gated recurrent unit (GRU) captures long-term dependence, an autoencoder (AE) downscales extracts features, multilayer perceptron (MLP) maps complex relationships. Subsequently, random forest (RF) integrates results obtain final prediction. The experimental show that mean absolute error (MAE) 0.32427, percentage (MAPE) 10.17428%, squared (MSE) 0.46626, root (RMSE) 0.68283, R-squared (R²) 93.10734%, accuracy rate (AR) 89.82572%. demonstrate proposed decomposition–selection–prediction–integration framework reduces errors, enhances stability through multiple experiments, improves efficiency accuracy, provides insights forecasting.

Language: Английский

---

Prediction OPEC oil price utilizing long short-term memory and multi-layer perceptron models

Alexandria Engineering Journal, Journal Year: 2024, Volume and Issue: 110, P. 607 - 612

Published: Oct. 18, 2024

Language: Английский

---

Hybrid price prediction method combining TCN-BiGRU and attention mechanism for battery-grade lithium carbonate

Kybernetes, Journal Year: 2024, Volume and Issue: unknown

Published: Sept. 11, 2024

Purpose To predict the price of battery-grade lithium carbonate accurately and provide proper guidance to investors, a method called MFTBGAM is proposed in this study. This integrates textual numerical information using TCN-BiGRU–Attention. Design/methodology/approach The Word2Vec model initially employed process gathered data concerning carbonate. Subsequently, dual-channel text-numerical extraction model, integrating TCN BiGRU, constructed extract features separately. Following this, attention mechanism applied fusion from data. Finally, market prediction results for are calculated outputted fully connected layer. Findings Experiments study carried out datasets consisting news investor commentary. findings reveal that exhibits superior performance compared alternative models, showing its efficacy precisely forecasting future Research limitations/implications dataset analyzed spans 2020 2023, thus, forecast specifically relevant timeframe. Altering sample would necessitate repetition experimental process, resulting different outcomes. Furthermore, recognizing raw might include noise irrelevant information, endeavors will explore efficient preprocessing techniques mitigate such issues, thereby enhancing model’s predictive capabilities long-term tasks. Social implications serves as valuable tool investors industry, facilitating informed investment decisions. By prediction, can discern opportune moments investment. Moreover, utilizes two distinct types text – comments independent sources input. approach provides with more precise comprehensive understanding dynamics. Originality/value We propose novel based on TCN-BiGRU Attention “text-numerical” fusion. separately use comments, enhance model's effectiveness generalization ability. Additionally, we utilize including both titles content improve accuracy predictions.

Language: Английский

---

An intelligent framework based on optimized variational mode decomposition and temporal convolutional network: Applications to stock index multi-step forecasting

Expert Systems with Applications, Journal Year: 2024, Volume and Issue: unknown, P. 126222 - 126222

Published: Dec. 1, 2024

Language: Английский

---