Neigbourhood Embedding for Improving Diffusion based Short Term Traffic Flow Prediction

Yin Hei Chan; Andrew Kwok Fai Lui; Sin Chun Ng

doi:10.1109/ISCAIE47305.2020.9108795

Neigbourhood Embedding for Improving Diffusion based Short Term Traffic Flow Prediction

Yin Hei Chan, Andrew Kwok Fai Lui, Sin Chun Ng

Hong Kong Metropolitan University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

1 Citation (Scopus)

Abstract

This paper describes a short-term traffic flow forecasting approach that combines efficient probability based diffusion mechanism and embedding of surrounding information to handle both ordinary situation and abnormal situations. A discrete diffusion model is insufficient in handle Spatial temporal traffic flow data sampled at regular time interval as there may be information loss during volatile environment. The hybrid model utilize DCRNN as discrete diffusion model and GRU as embedding, combining both for accurate prediction. Preliminary results shows improvement in both microscopic and macroscopic scales indicating the potential of hybrid approach towards accurate and efficient short term traffic flow forecasting.

Original language	English
Title of host publication	ISCAIE 2020 - IEEE 10th Symposium on Computer Applications and Industrial Electronics
Pages	186-191
Number of pages	6
ISBN (Electronic)	9781728150338
DOIs	https://doi.org/10.1109/ISCAIE47305.2020.9108795
Publication status	Published - Apr 2020
Event	10th IEEE Symposium on Computer Applications and Industrial Electronics, ISCAIE 2020 - Virtual, Malaysia Duration: 18 Apr 2020 → 19 Apr 2020

Publication series

Name	ISCAIE 2020 - IEEE 10th Symposium on Computer Applications and Industrial Electronics

Conference

Conference	10th IEEE Symposium on Computer Applications and Industrial Electronics, ISCAIE 2020
Country/Territory	Malaysia
City	Virtual
Period	18/04/20 → 19/04/20

Keywords

deep learning
diffusion convolution
gated recurrent unit
sequence to sequence model
traffic flow forecasting

Access to Document

10.1109/ISCAIE47305.2020.9108795

Cite this

Chan, Y. H., Lui, A. K. F., & Ng, S. C. (2020). Neigbourhood Embedding for Improving Diffusion based Short Term Traffic Flow Prediction. In ISCAIE 2020 - IEEE 10th Symposium on Computer Applications and Industrial Electronics (pp. 186-191). Article 9108795 (ISCAIE 2020 - IEEE 10th Symposium on Computer Applications and Industrial Electronics). https://doi.org/10.1109/ISCAIE47305.2020.9108795

@inproceedings{8ceb25afc7364d8a8126a8f346095efc,

title = "Neigbourhood Embedding for Improving Diffusion based Short Term Traffic Flow Prediction",

abstract = "This paper describes a short-term traffic flow forecasting approach that combines efficient probability based diffusion mechanism and embedding of surrounding information to handle both ordinary situation and abnormal situations. A discrete diffusion model is insufficient in handle Spatial temporal traffic flow data sampled at regular time interval as there may be information loss during volatile environment. The hybrid model utilize DCRNN as discrete diffusion model and GRU as embedding, combining both for accurate prediction. Preliminary results shows improvement in both microscopic and macroscopic scales indicating the potential of hybrid approach towards accurate and efficient short term traffic flow forecasting.",

keywords = "deep learning, diffusion convolution, gated recurrent unit, sequence to sequence model, traffic flow forecasting",

author = "Chan, {Yin Hei} and Lui, {Andrew Kwok Fai} and Ng, {Sin Chun}",

note = "Publisher Copyright: {\textcopyright} 2020 IEEE.; 10th IEEE Symposium on Computer Applications and Industrial Electronics, ISCAIE 2020 ; Conference date: 18-04-2020 Through 19-04-2020",

year = "2020",

month = apr,

doi = "10.1109/ISCAIE47305.2020.9108795",

language = "English",

series = "ISCAIE 2020 - IEEE 10th Symposium on Computer Applications and Industrial Electronics",

pages = "186--191",

booktitle = "ISCAIE 2020 - IEEE 10th Symposium on Computer Applications and Industrial Electronics",

}

Chan, YH, Lui, AKF & Ng, SC 2020, Neigbourhood Embedding for Improving Diffusion based Short Term Traffic Flow Prediction. in ISCAIE 2020 - IEEE 10th Symposium on Computer Applications and Industrial Electronics., 9108795, ISCAIE 2020 - IEEE 10th Symposium on Computer Applications and Industrial Electronics, pp. 186-191, 10th IEEE Symposium on Computer Applications and Industrial Electronics, ISCAIE 2020, Virtual, Malaysia, 18/04/20. https://doi.org/10.1109/ISCAIE47305.2020.9108795

Neigbourhood Embedding for Improving Diffusion based Short Term Traffic Flow Prediction. / Chan, Yin Hei; Lui, Andrew Kwok Fai; Ng, Sin Chun.
ISCAIE 2020 - IEEE 10th Symposium on Computer Applications and Industrial Electronics. 2020. p. 186-191 9108795 (ISCAIE 2020 - IEEE 10th Symposium on Computer Applications and Industrial Electronics).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Neigbourhood Embedding for Improving Diffusion based Short Term Traffic Flow Prediction

AU - Chan, Yin Hei

AU - Lui, Andrew Kwok Fai

AU - Ng, Sin Chun

PY - 2020/4

Y1 - 2020/4

N2 - This paper describes a short-term traffic flow forecasting approach that combines efficient probability based diffusion mechanism and embedding of surrounding information to handle both ordinary situation and abnormal situations. A discrete diffusion model is insufficient in handle Spatial temporal traffic flow data sampled at regular time interval as there may be information loss during volatile environment. The hybrid model utilize DCRNN as discrete diffusion model and GRU as embedding, combining both for accurate prediction. Preliminary results shows improvement in both microscopic and macroscopic scales indicating the potential of hybrid approach towards accurate and efficient short term traffic flow forecasting.

AB - This paper describes a short-term traffic flow forecasting approach that combines efficient probability based diffusion mechanism and embedding of surrounding information to handle both ordinary situation and abnormal situations. A discrete diffusion model is insufficient in handle Spatial temporal traffic flow data sampled at regular time interval as there may be information loss during volatile environment. The hybrid model utilize DCRNN as discrete diffusion model and GRU as embedding, combining both for accurate prediction. Preliminary results shows improvement in both microscopic and macroscopic scales indicating the potential of hybrid approach towards accurate and efficient short term traffic flow forecasting.

KW - deep learning

KW - diffusion convolution

KW - gated recurrent unit

KW - sequence to sequence model

KW - traffic flow forecasting

UR - http://www.scopus.com/inward/record.url?scp=85086628703&partnerID=8YFLogxK

U2 - 10.1109/ISCAIE47305.2020.9108795

DO - 10.1109/ISCAIE47305.2020.9108795

M3 - Conference contribution

AN - SCOPUS:85086628703

T3 - ISCAIE 2020 - IEEE 10th Symposium on Computer Applications and Industrial Electronics

SP - 186

EP - 191

BT - ISCAIE 2020 - IEEE 10th Symposium on Computer Applications and Industrial Electronics

T2 - 10th IEEE Symposium on Computer Applications and Industrial Electronics, ISCAIE 2020

Y2 - 18 April 2020 through 19 April 2020

ER -

Neigbourhood Embedding for Improving Diffusion based Short Term Traffic Flow Prediction

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this