Littérature scientifique sur le sujet « Dynaic prediction »

Purpose: This study compares models for predicting business financial crises, fo-cusing on which are most effective. In light of the new European Directive on business failure, it highlights a trade-off between predictive accuracy and timeli-ness in static models and offers an alternative approach. Design/methodology/approach: This study examines the Italian early warning system (EWS), testing static alert indicators' predictive ability on a large sample of private companies. It then proposes a dynamic version of the EWS. Findings: The results show a trade-off between predictive ability and timeliness for static models. In contrast, a dynamic system is more accurate in predicting cri-sis events, allowing managers to take corrective actions. Originality: The results highlight the limitations of static prediction models and emphasize the potential of a simple dynamic model that is specifically designed for small- and medium-sized entities (SMEs). Practical implications: This study proposes a dynamic model tailored for SMEs, which are particularly vulnerable to financial crises. This insight can help managers and policymakers balance accurate predictions with timely interventions, especial-ly in European countries implementing crisis prediction models.

Accurately and rapidly predicting marine drifter trajectories under conditions of information scarcity is critical for addressing maritime emergencies and conducting marine surveys with resource-limited unmanned vessels. Machine learning-based tracking methods, such as Long Short-Term Memory networks (LSTM), offer a promising approach for trajectory prediction in such scenarios. This study combines satellite observations and idealized simulations to compare the predictive performance of LSTM with a resource-dependent dynamic tracking method (DT). The results indicate that when driven solely by historical drifter paths, LSTM achieves better trajectory predictions when trained and tested on relative trajectory intervals rather than the absolute positions of individual trajectory points. In general, LSTM provides a more accurate geometric pattern of trajectories at the initial stages of forecasting, while DT offers superior accuracy in predicting specific trajectory positions. The velocity and curvature of ocean currents jointly influence the prediction quality of both methods. In regions characterized by active sub-mesoscale dynamics, such as the fast-flowing and meandering Kuroshio Current and Kuroshio Current Extension, DT predicts more reliable trajectory patterns but lacks precision in detailed position estimates compared to LSTM. However, in areas dominated by the fast but relatively straight North Equatorial Current, the performance of the two methods reverses. The two methods also demonstrate different tolerances for noise and sampling intervals. This study establishes a baseline for selecting machine learning methods for marine drifter prediction and highlights the limitations of AI-based predictions under data-scarce and resource-constrained conditions.

Social interactions involve processes ranging from face recognition to understanding others’ intentions. To guide appropriate behavior in a given context, social interactions rely on accurately predicting the outcomes of one's actions and the thoughts of others. Because social interactions are inherently dynamic, these predictions must be continuously adapted. The neural correlates of social processing have largely focused on emotion, mentalizing, and reward networks, without integration of systems involved in prediction. The cerebellum forms predictive models to calibrate movements and adapt them to changing situations, and cerebellar predictive modeling is thought to extend to nonmotor behaviors. Primary cerebellar dysfunction can produce social deficits, and atypical cerebellar structure and function are reported in autism, which is characterized by social communication challenges and atypical predictive processing. We examine the evidence that cerebellar-mediated predictions and adaptation play important roles in social processes and argue that disruptions in these processes contribute to autism.

Providing reliable predictive traffic information is a crucial element for successful operation of intelligent transportation systems. However, there are difficulties in providing accurate predictions mainly because of limitations in processing data associated with existing traffic surveillance systems and the lack of suitable prediction techniques. This study examines different aggregation intervals to characterize various levels of traffic dynamic representations and to investigate their effects on prediction accuracy. The relationship between data aggregation and predictability is explored by predicting travel times obtained from the inductive signature–based vehicle reidentification system on the I-405 freeway detector test bed in Irvine, California. For travel time prediction, this study employs three techniques: adaptive exponential smoothing, adaptive autoregressive model using Kalman filtering, and recurrent neural network with genetically optimized parameters. Finally, findings are discussed on suggestions for applying prediction techniques effectively.

Abstract. This paper addresses the use of the methods of nonlinear dynamics and chaos theory for building a predictive chaotic model from time series. The chaotic model predictions are made by the adaptive local models based on the dynamical neighbors found in the reconstructed phase space of the observables. We implemented the univariate and multivariate chaotic models with direct and multi-steps prediction techniques and optimized these models using an exhaustive search method. The built models were tested for predicting storm surge dynamics for different stormy conditions in the North Sea, and are compared to neural network models. The results show that the chaotic models can generally provide reliable and accurate short-term storm surge predictions.

Inverse dynamics from motion capture is the most common technique for acquiring biomechanical kinetic data. However, this method is time-intensive, limited to a gait laboratory setting, and requires a large array of reflective markers to be attached to the body. A practical alternative must be developed to provide biomechanical information to high-bandwidth prosthesis control systems to enable predictive controllers. In this study, we applied deep learning to build dynamical system models capable of accurately estimating and predicting prosthetic ankle torque from inverse dynamics using only six input signals. We performed a hyperparameter optimization protocol that automatically selected the model architectures and learning parameters that resulted in the most accurate predictions. We show that the trained deep neural networks predict ankle torques one sample into the future with an average RMSE of 0.04 ± 0.02 Nm/kg, corresponding to 2.9 ± 1.6% of the ankle torque’s dynamic range. Comparatively, a manually derived analytical regression model predicted ankle torques with a RMSE of 0.35 ± 0.53 Nm/kg, corresponding to 26.6 ± 40.9% of the ankle torque’s dynamic range. In addition, the deep neural networks predicted ankle torque values half a gait cycle into the future with an average decrease in performance of 1.7% of the ankle torque’s dynamic range when compared to the one-sample-ahead prediction. This application of deep learning provides an avenue towards the development of predictive control systems for powered limbs aimed at optimizing prosthetic ankle torque.

The burgeoning field of data analytics has emerged as a pivotal force in the realm of public health, particularly in the context of predicting and mitigating disease outbreaks. This comprehensive review delves into the diverse landscape of models and tools employed in data analytics for disease outbreak prediction. With a focus on synthesizing existing knowledge, the paper aims to provide a nuanced understanding of the strengths, limitations, and future directions within this dynamic field. The review begins with an exploration of various models utilized for disease outbreak prediction, ranging from statistical approaches to machine learning models and epidemiological frameworks. Each model category is scrutinized for its efficacy in capturing the complexities inherent in infectious disease dynamics. Simultaneously, the paper investigates the array of tools and technologies leveraged in disease outbreak prediction, encompassing Geographic Information Systems (GIS), data visualization tools, and big data analytics platforms. A critical aspect of the review lies in the examination of diverse data sources contributing to predictive analytics. Epidemiological data, environmental factors, and the burgeoning influence of social media and web data are dissected for their roles in enhancing the accuracy and timeliness of outbreak predictions. Amidst the promises of data analytics, the paper navigates the challenges inherent in predicting disease outbreaks. Issues of data quality and availability, model complexity, interpretability, and ethical considerations are dissected, providing a holistic view of the hurdles that practitioners encounter. Drawing upon case studies and real-world applications, the review showcases instances where data analytics has proven successful in predicting disease outbreaks, shedding light on both triumphs and setbacks. The implications for public health, lessons learned from challenges, and the evolving role of data analytics in shaping global health preparedness are thoroughly discussed. As the paper unfolds, it illuminates future trends and innovations in the field, foreseeing the integration of advanced technologies, global collaboration for information sharing, and the adaptation of predictive analytics for emerging diseases. The review culminates in a comprehensive conclusion, summarizing key findings and emphasizing the potential transformative impact of data analytics on the landscape of disease outbreak prediction.

The study focuses on predicting rental prices in Paris and aims to contribute to urban economics and data analytics. It analyzes a wide range of data sources, including historical rental prices, economic indicators, demographics, and regulations. The goal is to compare classical statistical models' prediction accuracy of these three models: ARIMA, dynamic regression, and random forest. The results reveal that the ARIMA model performs best, offering more accurate predictions. ARIMA relies on time series analysis, capturing complex patterns in rental prices, making it well-suited for dynamic real estate markets. The study also examines the impact of external factors like rent reference indices, house price indices, and unemployment rates on prediction accuracy. While these factors seem promising, the further analysis suggests they can introduce noise into predictions if not chosen carefully. In conclusion, this research contributes valuable insights for rental market stakeholders. The ARIMA model proves effective for rental price forecasting in Paris, emphasizing the importance of understanding intrinsic time series patterns. This study can guide decision-making for renters, landlords, and investors in Paris and similar urban areas, providing a better understanding of predictive modeling in housing markets.

Predictive analytics including statistical techniques, predictive modelling, machine learning, and data mining that analyse current and historical facts to make predictions about future or otherwise unknown events. Higher education institutions nowadays are under increasing pressure to respond to national and global economic, political and social changes such as the growing need to increase the proportion of students in certain disciplines, embedding workplace graduate attributes and ensuring that the quality of learning programs are both nationally and globally relevant. However, in higher education institution, there are significant numbers of students that stop their studies before graduation, especially for undergraduate students. Problem related to stopping out student and late or not graduating student can be improved by applying analytics. Using analytics, administrators, instructors and student can predict what will happen in future. Administrator and instructors can decide suitable intervention programs for at-risk students and before students decide to leave their study. Many different machine learning techniques have been implemented for predictive modelling in the past including decision tree, k-nearest neighbour, random forest, neural network, support vector machine, naïve Bayesian and a few others. A few attempts have been made to use Bayesian network and dynamic Bayesian network as modelling techniques for predicting at- risk student but a few challenges need to be resolved. The motivation for using dynamic Bayesian network is that it is robust to incomplete data and it provides opportunities for handling changing and dynamic environment. The trends and directions of research on prediction and identifying at-risk student are developing prediction model that can provide as early as possible alert to administrators, predictive model that handle dynamic and changing environment and the model that provide real-time prediction.

The COVID-19 pandemic is threatening our quality of life and economic sustainability. The rapid spread of COVID-19 around the world requires each country or region to establish appropriate anti-proliferation policies in a timely manner. It is important, in making COVID-19-related health policy decisions, to predict the number of confirmed COVID-19 patients as accurately and quickly as possible. Predictions are already being made using several traditional models such as the susceptible, infected, and recovered (SIR) and susceptible, exposed, infected, and resistant (SEIR) frameworks, but these predictions may not be accurate due to the simplicity of the models, so a prediction model with more diverse input features is needed. However, it is difficult to propose a universal predictive model globally because there are differences in data availability by country and region. Moreover, the training data for predicting confirmed patients is typically an imbalanced dataset consisting mostly of normal data; this imbalance negatively affects the accuracy of prediction. Hence, the purposes of this study are to extract rules for selecting appropriate prediction algorithms and data imbalance resolution methods according to the characteristics of the datasets available for each country or region, and to predict the number of COVID-19 patients based on these algorithms. To this end, a decision tree-type rule was extracted to identify 13 data characteristics and a discrimination algorithm was selected based on those characteristics. With this system, we predicted the COVID-19 situation in four regions: Africa, China, Korea, and the United States. The proposed method has higher prediction accuracy than the random selection method, the ensemble method, or the greedy method of discriminant analysis, and prediction takes very little time.