Unlock instant, AI-driven research and patent intelligence for your innovation.

A method to improve the accuracy of accident risk prediction for traffic dangerous persons

A technology for accident risk and prediction accuracy, applied in prediction, data processing applications, instruments, etc., can solve the problems of difficulty in improving model effect and complex integrated learning model, and achieve the effect of improving prediction accuracy and ensuring accuracy.

Active Publication Date: 2021-07-20
JIANGSU ZHITONG TRANSPORTATION TECH
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the complex parameter composition of the integrated learning model brings certain difficulties to the improvement of the model effect.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A method to improve the accuracy of accident risk prediction for traffic dangerous persons
  • A method to improve the accuracy of accident risk prediction for traffic dangerous persons
  • A method to improve the accuracy of accident risk prediction for traffic dangerous persons

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0043] A method to improve the accuracy of accident risk prediction for traffic dangerous persons, using an optimized sampling method to obtain traffic violation data and accident data samples, using integrated learning algorithms to train traffic accident risk prediction models for traffic participants, and further optimizing the model through genetic algorithms to improve The accuracy of prediction results, such as figure 1 . The method of the embodiment uses the integrated learning algorithm to mine the safety characteristics of traffic travelers in the traffic violation data, adopts the optimal sampling method in the sampling link of model construction to improve the performance of the initial model, and uses the genetic algorithm to optimize the model parameters to effectively improve the risk of accidents caused by high-risk personnel. Risk Prediction Accuracy. The specific method flow is:

[0044] S1. Based on the original traffic violation data and accident data, con...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The present invention provides a method for improving the prediction accuracy of accident risk of traffic dangerous persons, obtains traffic violation data and accident data samples with optimized sampling method, uses integrated learning algorithm to train traffic accident risk prediction model of traffic participants, and conducts model through genetic algorithm optimization. The invention uses an integrated learning algorithm to mine the safety characteristics of traffic travelers in the traffic violation data, adopts an optimized sampling method in the sampling link of model building to improve the performance of the initial model, and uses a genetic algorithm to optimize model parameters, effectively improving the accident risk of high-risk personnel prediction accuracy.

Description

technical field [0001] The invention relates to a method for improving the accident risk prediction accuracy of traffic dangerous persons. Background technique [0002] Studies have shown that there is a correlation between traffic violations and traffic accidents, and the attributes and behaviors of traffic participants such as drivers and pedestrians retained by traffic violations can provide data support for the analysis of human factors in traffic safety. Data mining can use the classification idea to mine the safety characteristics of traffic offenders according to the personnel attribute variables. [0003] The traditional classification method is to find a classifier that is closest to the actual classification function in a space composed of various possible functions, but in practice, only a weakly supervised model with preference is usually obtained, and the reliability of the model is not good. The ensemble learning algorithm improves the performance of the final...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06Q10/04G06Q50/26
CPCG06Q10/04G06Q50/26
Inventor 刘林陈凝吕伟韬马党生
Owner JIANGSU ZHITONG TRANSPORTATION TECH