Open AccessSegmentation of vehicle detector data for improved k‐nearest neighbours‐based traffic flow prediction
Author(s) -
Bernaś Marcin,
Płaczek Bartłomiej,
Porwik Piotr,
Pamuła Teresa
Publication year - 2015
Publication title -
iet intelligent transport systems
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.579
H-Index - 45
eISSN - 1751-9578
pISSN - 1751-956X
DOI - 10.1049/iet-its.2013.0164
Subject(s) - detector , traffic flow (computer networking) , segmentation , computer science , volume (thermodynamics) , traffic volume , data mining , basis (linear algebra) , k nearest neighbors algorithm , intelligent transportation system , term (time) , artificial intelligence , engineering , mathematics , transport engineering , telecommunications , computer security , physics , geometry , quantum mechanics
This study presents a data segmentation method, which was intended to improve the performance of the k‐nearest neighbours algorithm for making short‐term traffic volume predictions. According to the introduced method, selected segments of vehicle detector data are searched for records similar to the current traffic conditions, instead of the entire database. The data segments are determined on the basis of a segmentation procedure, which aims to select input data that are useful for the prediction algorithm. Advantages of the proposed method were demonstrated in experiments on real‐world traffic data. Experimental results show that the proposed method not only improves the accuracy of the traffic volume prediction, but also significantly reduces its computational cost.