Summary
Smart Info-Media Systems in Asia
2019
Session Number:RS1
Session:
Number:RS1-1
Research on Bus Data Pre-processing for Traffic Flow Prediction
Shuchen Gao, Hongye Yang, Jiaqi Zhang,
pp.67-71
Publication Date:2019/9/4
Online ISSN:2188-5079
Summary:
Traffic flow prediction is an important basis for intelligent transportation systems, and its accuracy directly affects traffic control and induced effects. In order to improve the accuracy of the prediction model and reduce the complexity of the model and lessen the training time, this article studies the pre-processing method of bus GPS data. First, the data correlation analysis shows that the key influencing factors are obtained, which lays the foundation for the selection of multiple input factors. Next, on the basis of ensuring the rationality of the data, the fusion algorithm based on SVDD and isolation forest is used to discriminate the outliers. Finally, a combined denoising algorithm based on Hanning+Symlet4 wavelet is proposed. Square error, signal noise ratio and smoothness are used as performance indicators to verify the effective optimization of the data, which provides powerful data support for further research.