文章基本信息

标题：A k Nearest Neighbor based Local Linear Wavelet Neural Network Model for On-line Short-term Traffic Volume Prediction
本地全文：下载
作者：Lei Lin ; Lei Lin ; Yan Li 等
期刊名称：Procedia - Social and Behavioral Sciences
印刷版ISSN：1877-0428
出版年度：2013
卷号：96
页码：2066-2077
DOI：10.1016/j.sbspro.2013.08.233
语种：English
出版社：Elsevier
摘要：AbstractThis paper presents a forecasting method called k nearest neighbor based local linear wavelet neural network (kNN-LLWNN) for the on-line, short-term prediction of five-minute traffic volumes at westbound of Interstate 64 in Hampton Road in Virginia. The method is based on combining k nearest neighbor (k-NN), with local linear wavelet neural network (LLWNN). The idea is to apply k-NN method to form the training dataset for LLWNN instead of taking the whole historical dataset for training. The proposed model is compared with k-NN, LLWNN and support vector regression (SVR) separately from prediction accuracy and running time two aspects. Experiments are conducted to decide the most appropriate parameters for the four models for the verification dataset. For the test dataset, the study's findings appear to confirm the hypothesis that, kNN-LLWNN performs comparable with LLWNN and SVR, and its running time is much lower than LLWNN and SVR because of the introduction of k-NN.
关键词：k nearest neighbor;local linear wavelet neural network;support vector regression;data item