A study of using artificial neural networks to develop an early warning predictor for credit union financial distress with comparison to the probit model

2001 ◽  
Vol 27 (4) ◽  
pp. 56-77 ◽  
Author(s):  
Clarence N.W. Tan ◽  
Herlina Dihardjo
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Early Warning ◽  
Financial Distress ◽  
Probit Model ◽  
Credit Union ◽  
Artificial Neural

scholarly journals Estimation of Rainfall from Climatology Data Using Artificial Neural Networks in Palembang City South Sumatera

2021 ◽  
Vol 930 (1) ◽  
pp. 012062
Author(s):  
E Suhartanto ◽  
S Wahyuni ◽  
K M Mufadhal
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Early Warning ◽  
Rainfall Data ◽  
Climatological Data ◽  
Data Requirement ◽  
Floods And Droughts ◽  
Artificial Neural

Abstract Estimation of climatological parameters, especially rainfall is a data requirement for all regions of Indonesia. The availability of rainfall data is used for early warning of flood or drought disasters. The study location is in Palembang City, South Sumatra Province, where floods and droughts often occur and lack of availability of rainfall data. This study aims to obtain the best model in estimating rainfall from climatological data. The analysis was carried out to estimate the rainfall from the climatological data using the Artificial Neural Networks method. The Artificial Neural Networks were applied and showed some results with the best calibration was at 16 years using TRAINLM with 1500 epochs that is the performances NSE = 0.54, RMSE = 99.37, and R = 0.74. Whereas the best validation was at 1 year that is the performances NSE = 0.41, RMSE = 87.32, and R = 0.65.

scholarly journals Brief communication: Rainfall thresholds based on Artificial neural networks can improve landslide early warning

2021 ◽  
Author(s):  
Pierpaolo Distefano ◽  
David J. Peres ◽  
Pietro Scandura ◽  
Antonino Cancelliere
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Performance Improvement ◽  
Early Warning ◽  
Rainfall Intensity ◽  
Rainfall Event ◽  
Rainfall Thresholds ◽  
Event Duration ◽  
Landslide Early Warning ◽  
Artificial Neural

Abstract. In this communication we show how the use of artificial neural networks (ANNs) can improve the performance of the rainfall thresholds for landslide early warning. Results for Sicily (Italy), show how performance of a traditional rainfall event duration and depth power law threshold, yielding a true skill statistic (TSS) of 0.50, can be improved by ANNs (TSS = 0.59). Then we show how ANNs allow to easily add other variables, like peak rainfall intensity, with a further performance improvement (TSS = 0.64). This may stimulate more research on the use of this powerful tool for deriving landslide early warning thresholds.

Sign in / Sign up

Export Citation Format

Share Document