Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Instance–Based Learning for Highway Accident Frequency Prediction
Accurate predictions of highway accident frequency may help traffic engineers design and test solutions for the improvement of highway safety. However, accident frequency prediction is by no means an easy task due to the large number of factors affecting accident occurrence and the complicated interactions among them. Many studies have been conducted to uncover the relationship between the roadway environment and corresponding accident frequencies. These studies used statistical approaches such as linear regression analysis. The actual relationship between the roadway environment and corresponding accident frequencies has not been approximated with an acceptable certainty because it usually coincided with the mathematical models assumed by the researchers. This paper describes an application of a machine learning method, instance–based learning (IBL), to highway accident frequency predictions. We developed an IBL system and applied this system to highway accident frequency predictions. The data set used contains accident data from the main Utah highways for a 5–year period (1988–1992). Experimental results show that the IBL method is applicable to highway accident predictions and compared favorably with linear regression analysis and neural networks.
Instance–Based Learning for Highway Accident Frequency Prediction
Accurate predictions of highway accident frequency may help traffic engineers design and test solutions for the improvement of highway safety. However, accident frequency prediction is by no means an easy task due to the large number of factors affecting accident occurrence and the complicated interactions among them. Many studies have been conducted to uncover the relationship between the roadway environment and corresponding accident frequencies. These studies used statistical approaches such as linear regression analysis. The actual relationship between the roadway environment and corresponding accident frequencies has not been approximated with an acceptable certainty because it usually coincided with the mathematical models assumed by the researchers. This paper describes an application of a machine learning method, instance–based learning (IBL), to highway accident frequency predictions. We developed an IBL system and applied this system to highway accident frequency predictions. The data set used contains accident data from the main Utah highways for a 5–year period (1988–1992). Experimental results show that the IBL method is applicable to highway accident predictions and compared favorably with linear regression analysis and neural networks.
Instance–Based Learning for Highway Accident Frequency Prediction
Zhang, Jianping (Autor:in) / Yang, Junming (Autor:in)
Computer‐Aided Civil and Infrastructure Engineering ; 12 ; 287-294
01.07.1997
8 pages
Aufsatz (Zeitschrift)
Elektronische Ressource
Englisch
Instance-Based Learning for Highway Accident Frequency Prediction
| Online Contents | 1997
NTIS | 1991
NTIS | 2003
NTIS | 1994