Air quality forecasting using SVR with quasi-linear kernel

Huilin Zhu, Jinglu Hu

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Air pollution has threaten people's health. It is urgent for the government to strengthen and enhance air pollution monitoring capacity. In this paper, we propose an air quality prediction model to infer air pollutant concentrations, such as CO, NOx, NO2. The idea is to design a sophisticate piecewise linear model by using a gated linear network. A top k% winner-take-all autoencoder is first built to generate a set of binary sequences as the gate control signals, so as to perform the input space partitioning. The piecewise linear model is then identified in an exact same way as a standard support vector regression (SVR) with a quasi-linear kernel composed by using the gate control signals. Results of our experiments shows that our proposed SVR prediction model outperforms other state-of-the-art methods.

Original languageEnglish
Title of host publicationCITS 2019 - Proceeding of the 2019 International Conference on Computer, Information and Telecommunication Systems
EditorsMohammad S. Obaidat, Zhenqiang Mi, Kuei-Fang Hsiao, Petros Nicopolitidis, Daniel Cascado-Caballero
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781538640883
DOIs
Publication statusPublished - 2019 Aug
Event2019 International Conference on Computer, Information and Telecommunication Systems, CITS 2019 - Beijing, China
Duration: 2019 Aug 282019 Aug 31

Publication series

NameCITS 2019 - Proceeding of the 2019 International Conference on Computer, Information and Telecommunication Systems

Conference

Conference2019 International Conference on Computer, Information and Telecommunication Systems, CITS 2019
CountryChina
CityBeijing
Period19/8/2819/8/31

Fingerprint

Air quality
Air pollution
Linear networks
Binary sequences
Health
Kernel
Support vector regression
Prediction model
Monitoring
Air
Experiments
Air pollutants
Winner-take-all
Government
Experiment
Partitioning
Top-k

Keywords

  • Air pollutant concentrations
  • Quasi-linear kernel
  • Support vector regression
  • Winner-take-all autoencoder

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition
  • Information Systems
  • Information Systems and Management
  • Safety, Risk, Reliability and Quality
  • Computer Networks and Communications
  • Hardware and Architecture

Cite this

Zhu, H., & Hu, J. (2019). Air quality forecasting using SVR with quasi-linear kernel. In M. S. Obaidat, Z. Mi, K-F. Hsiao, P. Nicopolitidis, & D. Cascado-Caballero (Eds.), CITS 2019 - Proceeding of the 2019 International Conference on Computer, Information and Telecommunication Systems [8862114] (CITS 2019 - Proceeding of the 2019 International Conference on Computer, Information and Telecommunication Systems). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/CITS.2019.8862114

Air quality forecasting using SVR with quasi-linear kernel. / Zhu, Huilin; Hu, Jinglu.

CITS 2019 - Proceeding of the 2019 International Conference on Computer, Information and Telecommunication Systems. ed. / Mohammad S. Obaidat; Zhenqiang Mi; Kuei-Fang Hsiao; Petros Nicopolitidis; Daniel Cascado-Caballero. Institute of Electrical and Electronics Engineers Inc., 2019. 8862114 (CITS 2019 - Proceeding of the 2019 International Conference on Computer, Information and Telecommunication Systems).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Zhu, H & Hu, J 2019, Air quality forecasting using SVR with quasi-linear kernel. in MS Obaidat, Z Mi, K-F Hsiao, P Nicopolitidis & D Cascado-Caballero (eds), CITS 2019 - Proceeding of the 2019 International Conference on Computer, Information and Telecommunication Systems., 8862114, CITS 2019 - Proceeding of the 2019 International Conference on Computer, Information and Telecommunication Systems, Institute of Electrical and Electronics Engineers Inc., 2019 International Conference on Computer, Information and Telecommunication Systems, CITS 2019, Beijing, China, 19/8/28. https://doi.org/10.1109/CITS.2019.8862114
Zhu H, Hu J. Air quality forecasting using SVR with quasi-linear kernel. In Obaidat MS, Mi Z, Hsiao K-F, Nicopolitidis P, Cascado-Caballero D, editors, CITS 2019 - Proceeding of the 2019 International Conference on Computer, Information and Telecommunication Systems. Institute of Electrical and Electronics Engineers Inc. 2019. 8862114. (CITS 2019 - Proceeding of the 2019 International Conference on Computer, Information and Telecommunication Systems). https://doi.org/10.1109/CITS.2019.8862114
Zhu, Huilin ; Hu, Jinglu. / Air quality forecasting using SVR with quasi-linear kernel. CITS 2019 - Proceeding of the 2019 International Conference on Computer, Information and Telecommunication Systems. editor / Mohammad S. Obaidat ; Zhenqiang Mi ; Kuei-Fang Hsiao ; Petros Nicopolitidis ; Daniel Cascado-Caballero. Institute of Electrical and Electronics Engineers Inc., 2019. (CITS 2019 - Proceeding of the 2019 International Conference on Computer, Information and Telecommunication Systems).
@inproceedings{35d951ae792647189f9d31b8ca400de5,
title = "Air quality forecasting using SVR with quasi-linear kernel",
abstract = "Air pollution has threaten people's health. It is urgent for the government to strengthen and enhance air pollution monitoring capacity. In this paper, we propose an air quality prediction model to infer air pollutant concentrations, such as CO, NOx, NO2. The idea is to design a sophisticate piecewise linear model by using a gated linear network. A top k{\%} winner-take-all autoencoder is first built to generate a set of binary sequences as the gate control signals, so as to perform the input space partitioning. The piecewise linear model is then identified in an exact same way as a standard support vector regression (SVR) with a quasi-linear kernel composed by using the gate control signals. Results of our experiments shows that our proposed SVR prediction model outperforms other state-of-the-art methods.",
keywords = "Air pollutant concentrations, Quasi-linear kernel, Support vector regression, Winner-take-all autoencoder",
author = "Huilin Zhu and Jinglu Hu",
year = "2019",
month = "8",
doi = "10.1109/CITS.2019.8862114",
language = "English",
series = "CITS 2019 - Proceeding of the 2019 International Conference on Computer, Information and Telecommunication Systems",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
editor = "Obaidat, {Mohammad S.} and Zhenqiang Mi and Kuei-Fang Hsiao and Petros Nicopolitidis and Daniel Cascado-Caballero",
booktitle = "CITS 2019 - Proceeding of the 2019 International Conference on Computer, Information and Telecommunication Systems",

}

TY - GEN

T1 - Air quality forecasting using SVR with quasi-linear kernel

AU - Zhu, Huilin

AU - Hu, Jinglu

PY - 2019/8

Y1 - 2019/8

N2 - Air pollution has threaten people's health. It is urgent for the government to strengthen and enhance air pollution monitoring capacity. In this paper, we propose an air quality prediction model to infer air pollutant concentrations, such as CO, NOx, NO2. The idea is to design a sophisticate piecewise linear model by using a gated linear network. A top k% winner-take-all autoencoder is first built to generate a set of binary sequences as the gate control signals, so as to perform the input space partitioning. The piecewise linear model is then identified in an exact same way as a standard support vector regression (SVR) with a quasi-linear kernel composed by using the gate control signals. Results of our experiments shows that our proposed SVR prediction model outperforms other state-of-the-art methods.

AB - Air pollution has threaten people's health. It is urgent for the government to strengthen and enhance air pollution monitoring capacity. In this paper, we propose an air quality prediction model to infer air pollutant concentrations, such as CO, NOx, NO2. The idea is to design a sophisticate piecewise linear model by using a gated linear network. A top k% winner-take-all autoencoder is first built to generate a set of binary sequences as the gate control signals, so as to perform the input space partitioning. The piecewise linear model is then identified in an exact same way as a standard support vector regression (SVR) with a quasi-linear kernel composed by using the gate control signals. Results of our experiments shows that our proposed SVR prediction model outperforms other state-of-the-art methods.

KW - Air pollutant concentrations

KW - Quasi-linear kernel

KW - Support vector regression

KW - Winner-take-all autoencoder

UR - http://www.scopus.com/inward/record.url?scp=85074157809&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85074157809&partnerID=8YFLogxK

U2 - 10.1109/CITS.2019.8862114

DO - 10.1109/CITS.2019.8862114

M3 - Conference contribution

AN - SCOPUS:85074157809

T3 - CITS 2019 - Proceeding of the 2019 International Conference on Computer, Information and Telecommunication Systems

BT - CITS 2019 - Proceeding of the 2019 International Conference on Computer, Information and Telecommunication Systems

A2 - Obaidat, Mohammad S.

A2 - Mi, Zhenqiang

A2 - Hsiao, Kuei-Fang

A2 - Nicopolitidis, Petros

A2 - Cascado-Caballero, Daniel

PB - Institute of Electrical and Electronics Engineers Inc.

ER -