Artificial neural networks have proved an attractive approach to non-linear regression problems arising in environmental modelling, such as statistical downscaling, short-term forecasting of atmospheric pollutant concentrations and rainfall run-off modelling. However, environmental datasets are frequently very noisy and characterized by a noise process that may be heteroscedastic (having input dependent variance) and/or non-Gaussian. The aim of this paper is to review existing methodologies for estimating predictive uncertainty in such situations and, more importantly, to illustrate how a model of the predictive distribution may be exploited in assessing the possible impacts of climate change and to improve current decision making processes. The results of the WCCI-2006 predictive uncertainty in environmental modelling challenge are also reviewed, suggesting a number of areas where further research may provide significant benefits.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1016/j.neunet.2007.04.024 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Prognostic value of CHA2DS2-VASc score for in-hospital outcomes in patients with Takotsubo syndrome.
Cardiovasc Revasc Med
December 2024
Division of Cardiology, Department of Medicine, Warren Alpert Medical School of Brown University and Lifespan Cardiovascular Institute, Providence, RI, USA.
Background: There is uncertainty about the use of the CHA2DS2-VASc score to predict clinical events in patients with Takotsubo syndrome (TTS). This study aimed to assess the short-term prognostic role of CHA2DS2-VASc score in this population.
Methods: All admissions with a primary diagnosis of TTS were included using data from the National Inpatient Sample database during 2016-2019.
Long-term forecast for antibacterial drug consumption in Germany using ARIMA models.
Naunyn Schmiedebergs Arch Pharmacol
January 2025
Hannover Medical School, Institute of Pharmacology, D-30625, Hannover, Germany.
The increasing supply shortages of antibacterial drugs presents significant challenges to public health in Germany. This study aims to predict the future consumption of the ten most prescribed antibacterial drugs in Germany up to 2040 using ARIMA (Auto Regressive Integrated Moving Average) models, based on historical prescription data. This analysis also evaluates the plausibility of the forecasts.
View Article and Find Full Text PDFHome Urine Dipstick Screening for Bladder and Kidney Cancer in High-Risk Populations in England: A Microsimulation Study of Long-Term Impact and Cost-Effectiveness.
Pharmacoeconomics
January 2025
Sheffield Centre for Health and Related Research (SCHARR), School of Medicine and Population Health, The University of Sheffield, Regent Court, 30 Regent Street, Sheffield, UK.
Background: Testing high-risk populations for non-visible haematuria may enable earlier detection of bladder cancer, potentially decreasing mortality. This research aimed to assess the cost-effectiveness of urine dipstick screening for bladder cancer in high-risk populations in England.
Methods: A microsimulation model developed in R software was calibrated to national incidence data by age, sex and stage, and validated against mortality data.
A novel fuzzy three-valued logic computational framework in machine learning for medicine dataset.
Comput Biol Med
January 2025
Department of Mathematics, NED University of Engineering & Technology, Pakistan. Electronic address:
For consideration of uncertainties of a medicine dataset, a new conceptual architecture fuzzy three-valued logic is introduced in this research work. The proposed concept is applied to the heart disease dataset for the assessment of heart disease risk in individuals. By comparison of three binary (0,1) input variables, the variables' uncertainties and their collective impact can be analyzed that provide complete information leading to better outcome prediction.
View Article and Find Full Text PDFDemystifying the art of isotope-enabled hydrological and climate modelling.
Sci Total Environ
January 2025
University of Tokyo, Japan.
Over the last 20 years, we have dramatically improved hydrometeorological data including isotopes, but are we making the most of this data? Stable isotopes of oxygen and hydrogen in the water molecule (stable water isotopes - SWI) are well known tracers of the global hydrological cycle producing critical climate science. Despite this, stable water isotopes are not explicitly included in influential climate reports (e.g.
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!