期刊
JOURNAL OF BIG DATA
卷 5, 期 1, 页码 -出版社
SPRINGERNATURE
DOI: 10.1186/s40537-018-0157-0
关键词
Traffic density; Multilayer perceptron with back-propagation; Multi-class traffic; Statistical indices; Traffic flow
资金
- Ministry of Education and Science of the Russian Federation [4-2017-052]
Short term traffic forecasting is one of the important fields of study in the transportation domain. Short term traffic forecasting is very useful to develop a more advanced transportation system to control traffic signals and avoid congestions. Several studies have made efforts for short term traffic flow forecasting for divided and undivided highways across the world. However, all these studies relied on the dataset which are greatly varied between countries due to the technology used for transportation data collection. India is a developing country in which efforts are being done to improve the transportation system to avoid congestion and travel time. Two-lane undivided highways with mixed traffic constitute a large portion of Indian road network. This study is an attempt to develop a short term traffic forecasting model using back propagation artificial neural network for two lane undivided highway with mixed traffic conditions in India. The results were compared with random forest, support vector machine, k-nearest neighbor classifier, regression tree and multiple regression models. It was found that back-propagation neural network performs better than other approaches and achieved an R-2 value 0.9962, which is a good score.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据