Home

Aims and Scope

Instructions for Authors

View Issues & Articles

Editorial Board

Article Search

ATS International Journal
Editor in Chief: Prof. Alessandro Calvi
Address: Via Vito Volterra 62,
00146, Rome, Italy.
Mail to: alessandro.calvi@uniroma3.it

Perception of driving safety risks at urban intersections based on multidimensional data mining

L. Yao, X.L. Cao
Pages: 111-124

Abstract:

Research on the perception of driving safety risks holds significant importance for enhancing traffic safety and facilitating the development of smart cities. A new perception method of driving safety risks at urban intersections based on multidimensional data mining is proposed to solve the problems of low accuracy, low recall, and long time consumption of traditional methods. Mining multidimensional urban intersection data, smoothing the data using Exponential Weighted Moving Average (EWMA) method, and repairing missing values using Gradient Boosting Tree (GBDT). Using the Apriori algorithm to mine association rules and determine key factors affecting driving safety, a perception index system is established. The perception index is used as input parameters to build a BP neural network, thereby achieving driving safety risk perception. The experimental results show that the accuracy of the proposed method is over 95%, the recall rate is over 90%, and it takes less than 3s.
Keywords: multidimensional data mining; urban intersection; driving safety; risk perception; apriori algorithm

2025 ISSUES
2024 ISSUES
2023 ISSUES
2022 ISSUES
2021 ISSUES
2020 ISSUES
2019 ISSUES
2018 ISSUES
2017 ISSUES
2016 ISSUES
2015 ISSUES
2014 ISSUES
2013 ISSUES
2012 ISSUES
2011 ISSUES
2010 ISSUES
2009 ISSUES
2008 ISSUES
2007 ISSUES
2006 ISSUES
2005 ISSUES
2004 ISSUES
2003 ISSUES