Comparative study of multivariate hybrid neural networks for global sea level prediction through 2050

Environmental Earth Sciences, Journal Year: 2025, Volume and Issue: 84(3)

Published: Jan. 21, 2025

Language: Английский

---

A Deep Learning-Based Ensemble Framework to Predict IPOs Performance for Sustainable Economic Development

Sustainability, Journal Year: 2025, Volume and Issue: 17(3), P. 827 - 827

Published: Jan. 21, 2025

Addressing resource scarcity and climate change necessitates a transition to sustainable consumption circular economy models, fostering environmental, social, economic resilience. This study introduces deep learning-based ensemble framework optimize initial public offering (IPO) performance prediction while extending its application processes, such as recovery waste reduction. The incorporates advanced techniques, including hyperparameter optimization, dynamic metric adaptation (DMA), the synthetic minority oversampling technique (SMOTE), address challenges class imbalance, risk-adjusted enhancement, robust forecasting. Experimental results demonstrate high predictive performance, achieving an accuracy of 76%, precision 83%, recall 75%, AUC 0.9038. Among methods, Bagging achieved highest (0.90), outperforming XGBoost (0.88) random forest (0.75). Cross-validation confirmed framework’s reliability with median 0.85 across ten folds. When applied scenarios, model effectively predicted sustainability metrics, R² values 0.76 for both reduction low mean absolute error (MAE = 0.11). These highlight potential align financial forecasting environmental objectives. underscores transformative learning in addressing challenges, demonstrating how AI-driven models can integrate goals. By enabling IPO predictions enhancing outcomes, proposed aligns Industry 5.0’s vision human-centric, data-driven, industrial innovation, contributing resilient growth long-term stewardship.

Language: Английский

---

Hybrid CNN-BiGRU-AM Model with Anomaly Detection for Nonlinear Stock Price Prediction

Electronics, Journal Year: 2025, Volume and Issue: 14(7), P. 1275 - 1275

Published: March 24, 2025

To address challenges in stock price prediction including data nonlinearity and anomalies, we propose a hybrid CNN-BiGRU-AM framework integrated with deep learning-based anomaly detection. First, an detection module identifies irregularities data. The CNN component then extracts local features while filtering anomalous information, followed by nonlinear pattern modeling through BiGRU attention mechanisms. Final predictions undergo secondary screening to ensure reliability. Experimental evaluation on Shanghai Composite (SSE) daily closing prices demonstrates superior performance R2 = 0.9903, RMSE 22.027, MAE 19.043, Sharpe Ratio of 0.65. It is noteworthy that the this model reduced 14.7%, decreased 7.7% compared its ablation model. achieves multi-level feature extraction convolutional operations bidirectional temporal modeling, effectively enhancing generalization via mapping correction. Comparative analysis across models provides practical insights for investment decision-making. This dual-functional system not only improves accuracy but also offers interpretable references market mechanism regulatory policy formulation.

Language: Английский

---

Mining Frequent Sequences with Time Constraints from High-Frequency Data

International Journal of Financial Studies, Journal Year: 2025, Volume and Issue: 13(2), P. 55 - 55

Published: April 3, 2025

Investing in the stock market has always been an exciting topic for people. Many specialists have tried to develop tools predict future prices order make high profits and avoid big losses. However, predicting based on dynamic characteristics of stocks seems be a non-trivial problem. In practice, predictive models are not expected provide most accurate forecasts prices, but highlight changes discrepancies between predicted observed values, warn against threats, inform users about upcoming opportunities. this paper, we discuss use frequent sequences as well association rules WIG20 price prediction. Specifically, our study used two methods approach problem: correlation analysis Pearson coefficient sequence mining with temporal constraints. total, 43 were discovered, characterized by relatively confidence lift. Moreover, effective those that described same type trend both companies, i.e., rise ⇒ rise, or fall fall. showed opposite trend, namely fall, rare.

Language: Английский

---

Exploring Machine Learning for Stock Price Prediction and Decision Making

Published: April 19, 2025

Intricate dynamics of the stock market makes its prediction a challenging and daunting activity. In order to create precise predictive models, researchers are employing emerging machine learning models methods. The research starts with collection history, volumes trade other related indicators. Then data is preprocessed feature engineering done, thereby producing useful input representations for models. model employed in SVR model. Grid search CV method utilized discover best possible parameters' values that assists predicting intraday based on recent past data. This respond promptly trends changes, making it optimal short-term momentum trading strategies.

Language: Английский

---

A Fuzzy Decision Support System for Real Estate Valuations

Electronics, Journal Year: 2024, Volume and Issue: 13(24), P. 5046 - 5046

Published: Dec. 22, 2024

The field of real estate valuations is multivariate in nature. Each property has different intrinsic attributes that have a bearing on its final value: location, use, purpose, access, the services available to it, etc. appraiser analyzes all these factors and current status other similar properties market (comparable assets or units comparison) subjectively, with no applicable rules metrics, obtain value question. To model this context subjectivity, paper proposes use fuzzy system. inputs system designed are variables considered by appraiser, output adjustment coefficient be applied price each comparable asset appraised. design model, data been extracted from actual appraisals conducted three professional appraisers urban center Santa Cruz de Tenerife (Canary Islands, Spain). decision-helping tool sector: can it select most suitable comparables automatically coefficients, freeing them arduous task calculating manually based multiple parameters consider. Finally, an evaluation presented demonstrates applicability.

Language: Английский

---