scholarly journals A Comparative Analysis and Prediction of Traffic Accident Causalities in the Sultanate of Oman using Artificial Neural Networks and Statistical methods

1998 ◽  
Vol 3 ◽  
pp. 11 ◽  
Author(s):  
Galal A. Ali ◽  
Saleh M. Al-Alawi ◽  
Charles S.Bakheit
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Traffic Accident ◽  
Traffic Accidents ◽  
Regression Equations ◽  
Sultanate Of Oman ◽  
Related Factors ◽  
Population Number ◽  
Artificial Neural ◽  
On The Road

Traffic accidents are among the major causes of death in the Sultanate of Oman This is particularly the case in the age group of I6 to 25. Studies indicate that, in spite of Oman's high population-per-vehicle ratio, its fatality rate per l0,000 vehicles is one of the highest in the world. This alarming Situation underlines the importance of analyzing traffic accident data and predicting accident casualties. Such steps will lead to understanding the underlying causes of traffic accidents, and thereby to devise appropriate measures to reduce the number of car accidents and enhance safety standards. In this paper, a comparative study of car accident casualties in Oman was undertaken. Artificial Neural Networks (ANNs) were used to analyze the data and make predictions of the number of accident casualties. The results were compared with those obtained from the analysis and predictions by regression techniques. Both approaches attempted to model accident casualties using historical  data on related factors, such as population, number of cars on the road and so on, covering the period from I976 to 1994. Forecasts for the years 1995 to 2000 were made using ANNs and regression equations. The results from ANNs provided the best fit for the data. However, it was found that ANNs gave lower forecasts relative to those obtained by the regression methods used, indicating that ANNs are suitable for interpolation but their use for extrapolation may be limited. Nevertheless, the study showed that ANNs provide a potentially powerful tool in analyzing and forecasting traffic accidents and casualties.

scholarly journals Prediction of Carcass Meat Percentage in Young Pigs Using Linear Regression Models and Artificial Neural Networks

2016 ◽  
Vol 16 (1) ◽  
pp. 275-286 ◽  
Author(s):  
Magdalena Szyndler-Nędza ◽  
Robert Eckert ◽  
Tadeusz Blicharski ◽  
Mirosław Tyra ◽  
Artur Prokowski
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Linear Regression ◽  
Regression Model ◽  
Linear Regression Model ◽  
Performance Testing ◽  
Regression Equations ◽  
Young Pigs ◽  
Marquardt Algorithm ◽  
Artificial Neural

Abstract One of the approaches to improving performance testing of pigs is to look for mathematical solutions to increase the accuracy of calculations. This is mainly done through improvement of linear regression equations based on current data on performance tested pigs in Poland. The advances in computer technology and the improvements in mathematical analysis have made it possible to use artificial neural networks (ANNs) for prediction of carcass meat percentage in young pigs. The aim of the study was to compare the potential for live estimation of carcass meat percentage in pigs using two computational methods: linear regression equations and ANNs. The experiment used 654 gilts of six breeds, which were subjected to performance testing and slaughter analysis at the Pig Performance Testing Station (SKURTCh). The collected data were used to train ANNs to estimate carcass meat percentage in young pigs. Training was performed using the Levenberg- Marquardt algorithm. Next, meatiness estimated by ANNs was compared with the results obtained using linear modelling. It is concluded that based on the fattening and slaughter performance test results of live pigs, artificial neural networks (SSN23) are significantly more accurate in estimating carcass meat percentage in young pigs compared to the three-variable linear regression model 1. The difference in meatiness estimation between SSN23 and the four-variable linear regression model 2 was statistically non-significant in most of the breeds except Duroc and Pietrain, where the meatiness of young animals was estimated more accurately by the linear regression model.

Effect of Noisy Data on Pavement Performance Prediction by Artificial Neural Networks

1998 ◽  
Vol 1643 (1) ◽  
pp. 7-13 ◽  
Author(s):  
A. Raja Shekharan
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Noisy Data ◽  
Pavement Performance ◽  
Regression Equations ◽  
Predictive Capability ◽  
The Neural Networks ◽  
Artificial Neural ◽  
Pavement Performance Prediction ◽  
The Relationship

Artificial neural networks are increasingly employed in prediction modeling and are particularly advantageous when the relationship between the response and the predictor variables is complex. For the purposes of prediction, neural networks are to be trained with data that are accurately compiled. Frequently, the data collected either from field or laboratory observations are noisy in nature. The effect of noisy data on the predictive capability of neural networks has been studied. Present serviceability rating (PSR) of pavements is the attribute to be predicted. Six noisy databases are created and are employed to train the neural networks to predict PSR. Regression equations are developed with the same noisy databases, and the predictions from neural networks are compared with those of regression. The results show that the neural networks predict PSR as accurately as regression models with a given noisy data. In addition, neural networks are trained with data containing no noise. If no noise is present in the data, neural networks predict PSR accurately while properly capturing the effect of each explanatory variable on the response variable.

scholarly journals Application Of L1- Regularization Approach In QSAR Problem. Linear Regression And Artificial Neural Networks

2019 ◽  
Vol 14 (2) ◽  
pp. 79-90
Author(s):  
M.I. Berdnyk ◽  
A.B. Zakharov ◽  
V.V. Ivanov
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Least Squares ◽  
Principal Component Regression ◽  
Principal Component ◽  
Ordinary Least Squares ◽  
Regression Equations ◽  
Absolute Deviation ◽  
Internal Validation ◽  
Artificial Neural

One of the primary tasks of analytical chemistry and QSAR/QSPR researches is building of prognostic regression equations based on descriptors sets. The one of the most important problems here is to decrease the number of descriptors in the initial descriptor set which is usually way too big. In current investigation the descriptor set is proposed to be reduced employing the least absolute shrinkage and selection operator (LASSO) approach. Decreased descriptor sets were used for calculations with application of the following QSAR/QSPR methods: ordinary least squares (OLS), the least absolute deviation (LAD) regressions and artificial neural networks (ANN). Contrary to aforementioned methods principal component regression (PCR) and partial least squares (PLS) approaches can produce solutions containing numerous descriptors. In this article we compared the viability of these two different descriptor handling ideologies in application to molecular chemical and physical properties prediction. From the obtained results it is possible to see that there are tasks for which PCR and PLS approaches can fail to produce accurate regression equations. At the same time, methods OLS and LAD that use small amount of descriptors can provide viable solutions for the same cases. It was shown that these small sets of descriptors selected with LASSO approach can be used in ANN to obtain models with even better internal validation characteristics.

Sign in / Sign up

Export Citation Format

Share Document