DocumentCode
501826
Title
Application Research on Machine Learning and Statistical Forecasting Algorithm in Traffic Information Forecasting System
Author
Pei, Yan ; Yang, Guang-Ming
Author_Institution
Software Sch., Northeastern Univ., Shenyang, China
Volume
2
fYear
2009
fDate
12-14 Aug. 2009
Firstpage
431
Lastpage
436
Abstract
This paper proposes an algorithm model based on the machine learning and statistic forecasting method to forecast the traffic information. It constructs the forecasting specimens set by the mechanism learning of the machine learning, forecasts the short time traffic information by the data fit return technology, and constructs the knowledge base, which uses the concept of knowledge maturation degree to judge the application probability rate of the forecasted information in the traffic information forecasting system. It implements the short time traffic flowing forecasting in the traffic information forecasting system by this algorithm model, and it gets the effective to forecast the change of the traffic flowing state on some routes or transport corridor in the few minutes of the future.
Keywords
knowledge based systems; learning (artificial intelligence); probability; regression analysis; set theory; traffic information systems; data fit return technology; knowledge base construction; knowledge maturation degree; machine learning; probability rate; regression algorithm model; statistical specimen set forecasting algorithm; traffic flow information forecasting system; transport corridor; Communication system traffic control; Machine learning; Machine learning algorithms; Predictive models; Radio navigation; Road vehicles; Software algorithms; Technology forecasting; Telecommunication traffic; Traffic control; data fit; knowledge maturation degree; mechanism learning; traffic information forecasting;
fLanguage
English
Publisher
ieee
Conference_Titel
Hybrid Intelligent Systems, 2009. HIS '09. Ninth International Conference on
Conference_Location
Shenyang
Print_ISBN
978-0-7695-3745-0
Type
conf
DOI
10.1109/HIS.2009.202
Filename
5254500
Link To Document