期刊名称:International Journal of Advanced Computer Science and Applications(IJACSA)
印刷版ISSN:2158-107X
电子版ISSN:2156-5570
出版年度:2020
卷号:11
期号:12
页码:762-770
DOI:10.14569/IJACSA.2020.0111287
出版社:Science and Information Society (SAI)
摘要:Road safety researchers working on road accident data have witnessed success in road traffic accidents analysis through the application data analytic techniques, though, little progress was made into the prediction of road injury. This paper applies advanced data analytics methods to predict injury severity levels and evaluates their performance. The study uses predictive modelling techniques to identify risk and key factors that contributes to accident severity. The study uses publicly available data from UK department of transport that covers the period from 2005 to 2019. The paper presents an approach which is general enough so that can be applied to different data sets from other countries. The results identified that tree based techniques such as XGBoost outperform regression based ones, such as ANN. In addition to the paper, identifies interesting relationships and acknowledged issues related to quality of data.
关键词:Traffic Accidents Analytics (RTA); data mining; machine learning; XGBOOST