A Comparative Study on Deep Learning Models for Stock Price Prediction

Advances in transdisciplinary engineering, Год журнала: 2024, Номер unknown

Опубликована: Июль 1, 2024

With the advancement of deep learning technology, researchers have begun employing neural network models such as LSTM, MLP, CNN, and GRU to tackle nonlinear prediction problems in stock markets. This study harnesses these models, including GRU, investigate forecasting market price indices. The experiment selects globally representative indices and, through analyzing short-term, mid-term, long-term data, discovers that model demonstrates exceptional performance both accuracy generalization capabilities. To further enhance precision, research constructs a filtering ensemble decision-making system based on multiple models. strategy reduces error by at least 0.3% medium predictions, significantly outperforming single standalone model. Consequently, GRU-filtered decision has been empirically proven an effective tool for predicting indices, not only broadening methods available but also deepening understanding application abilities within financial

Язык: Английский

Optimization Techniques for Asthma Exacerbation Prediction Models: A Systematic Literature Review

IEEE Access, Год журнала: 2024, Номер 12, С. 110862 - 110890

Опубликована: Янв. 1, 2024

Asthma exacerbations pose a significant global health concern, necessitating effective predictive models to anticipate and manage these events. This systematic literature review examined the optimization techniques employed in asthma exacerbation prediction models, spanning machine learning algorithms computational methods. The objective was synthesize existing evidence, identify trends, delineate future research directions modeling for enhance accuracy clinical utility. A comprehensive search strategy devised, yielding 27 eligible articles analysis. result revealed various techniques, including feature selection, model optimization, environmental factor integration. also that algorithms' effectiveness predicting varied depending on factors (such as dataset quality complexity), with selection ensemble learning) used improving accuracy. Integrating spatial enhanced enabling tailored interventions. In addition, personalized management strategies informed by led better control reduced healthcare utilization. highlighted implications management, well methodological limitations, proposed improve reliability advance understanding, thereby contributing United Nations' Sustainable Development Goals related health, innovation, sustainability. Thus, progress made identification of challenges areas improvement were covered, providing valuable insights researchers, clinicians, policymakers aiming care through modeling.

Язык: Английский