交通流数据的概念漂移探析
A Brief Discussion on Conceptual Drift of Traffic Flow Data
李玲玲 1辛浩1
作者信息
- 1. 淮北职业技术学院计算机科学技术系,安徽淮北 235000
- 折叠
摘要
概念漂移在很大程度上影响着交通流预测的结果,当数据分布或属性发生了变化,而模型的参数和结构没有及时调整时,模型预测结果的准确性就会大幅度下降,甚至完全失效.研究交通流数据及特点,概念漂移及检测步骤、检测方法,对交通流数据概念漂移发现和检测模型的建立、后期调整和优化交通流预测模型具有重要意义.
Abstract
Conceptual drift greatly affects the results of traffic flow prediction.When the data distri-bution or attributes change,and the parameters and structure of the model are not adjusted in a timely manner,the accuracy of the model's prediction results will significantly decrease,and even become com-pletely ineffective.Studying traffic flow data and its characteristics,conceptual drift and its detection steps and methods is of great significance for the establishment of traffic flow data conceptual drift detec-tion and detection models,as well as timely adjustment and optimization of traffic flow prediction models in the later stage.
关键词
交通流数据/概念漂移/检测Key words
traffic flow data/conceptual drift/detection引用本文复制引用
基金项目
安徽省自然科学研究重点项目(2023AH052869)
淮北职业技术学院自然科学研究重点项目(2020-A-6)
淮北职业技术学院自然科学研究项目(2021-A-2)
出版年
2024
NETL
NSTL
维普
万方数据