This repo is for the paper:

Short-Term Traffic Forecasting Using Self-Adjusting k-Nearest Neighbours.

It is about short-term prediction by self-adjusting dynamic procedure kNN (DP-kNN).

Published by Journal (Abbr.): IET Intell. Transp. Syst.

DOI: 10.1049/iet-its.2016.0263.

Some content will be updated later.

Abstract

Short-term traffic forecasting is becoming more important in intelligent transportation systems. The k-nearest neighbours (kNN) method is widely used for short-term traffic forecasting. However, kNN parameters self-adjustment has been a problem due to dynamic traffic characteristics. This paper proposes a fully automatic dynamic procedure kNN (DP-kNN) that makes the kNN parameters self-adjustable and robust without predefined models or training. We used realworld data with more than one-year traffic records to conduct experiments. The results show that DP-kNN can perform better than manually adjusted kNN and other benchmarking methods with regards to accuracy on average. This study also discusses the difference between holiday and workday traffic prediction as well as the usage of neighbour distance measurement.

Code

Level 1 code is available in repo: level 1, which is used to generate parameters' combinations' results. Level 2 code is available in folder R. Note: the implementation is not considering the data as stream as it just make the handling of data non-necessaryly more complex than evaluation's requirement, more like an industry software.

Results

The results show that DP-kNN gives 9% to 40% improvement than benchmarking methods on average.

Citation Request

[IEEE Format] B. Sun, W. Cheng, P. Goswami, and G. Bai, “Short-Term Traffic Forecasting Using Self-Adjusting k-Nearest Neighbours,” IET Intelligent Transport Systems, 2018.

[AAA Format] Bin Sun, Wei Cheng, Prashant Goswami, and Guohua Bai, 2018. Short-Term Traffic Forecasting Using Self-Adjusting K-Nearest Neighbours. IET Intelligent Transport Systems.

[GB/T 7714] SUN Bin, CHENG Wei, GOSWAMI Prashant, et al. Short-Term Traffic Forecasting Using Self-Adjusting k-Nearest Neighbours[J]. IET Intelligent Transport Systems, 2018.

[Bibtex]:

@article{sun2018shortterm,
  title = {Short-{{Term Traffic Forecasting Using Self}}-{{Adjusting}} k-{{Nearest Neighbours}}},
  issn = {1751-956X},
  doi = {10.1049/iet-its.2016.0263},
  abstract = {Short-term traffic forecasting is becoming more important in intelligent transportation systems. The k-nearest neighbours (kNN) method is widely used for short-term traffic forecasting. However, kNN parameters self-adjustment has been a problem due to dynamic traffic characteristics. This paper proposes a fully automatic dynamic procedure kNN (DP-kNN) that makes the kNN parameters self-adjustable and robust without predefined models or training. We used realworld data with more than one-year traffic records to conduct experiments. The results show that DP-kNN can perform better than manually adjusted kNN and other benchmarking methods with regards to accuracy on average. This study also discusses the difference between holiday and workday traffic prediction as well as the usage of neighbour distance measurement.},
  journaltitle = {IET Intelligent Transport Systems},
  author = {Sun, Bin and Cheng, Wei and Goswami, Prashant and Bai, Guohua},
  date = {2018},
}

Paper

The paper full-text is available on IET Digital Library. The code is available on GitHub.