Improved Interpolation and Anomaly Detection for Personal PM<sub>2.5</sub> Measurement
With the development of technology, especially technologies related to artificial intelligence (AI), the fine-dust data acquired by various personal monitoring devices is of great value as training data for predicting future fine-dust concentrations and innovatively alerting people of potential dang...
Main Authors: | , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2020-01-01
|
Series: | Applied Sciences |
Subjects: | |
Online Access: | https://www.mdpi.com/2076-3417/10/2/543 |
id |
doaj-49da864a199a415bbf1323c6589ec22d |
---|---|
record_format |
Article |
spelling |
doaj-49da864a199a415bbf1323c6589ec22d2020-11-25T00:33:27ZengMDPI AGApplied Sciences2076-34172020-01-0110254310.3390/app10020543app10020543Improved Interpolation and Anomaly Detection for Personal PM<sub>2.5</sub> MeasurementJinSoo Park0Sungroul Kim1Department of Industrial Cooperation, Soonchunhyang University, Asan 31538, KoreaDepartment of Environmental Sciences, Soonchunhyang University, Asan 31538, KoreaWith the development of technology, especially technologies related to artificial intelligence (AI), the fine-dust data acquired by various personal monitoring devices is of great value as training data for predicting future fine-dust concentrations and innovatively alerting people of potential danger. However, most of the fine-dust data obtained from those devices include either missing or abnormal data caused by various factors such as sensor malfunction, transmission errors, or storage errors. This paper presents methods to interpolate the missing data and detect anomalies in PM<sub>2.5</sub> time-series data. We validated the performance of our method by comparing ours to well-known existing methods using our personal PM<sub>2.5</sub> monitoring data. Our results showed that the proposed interpolation method achieves more than 25% improved results in root mean square error (RMSE) than do most existing methods, and the proposed anomaly detection method achieves fairly accurate results even for the case of the highly capricious fine-dust data. These proposed methods are expected to contribute greatly to improving the reliability of data.https://www.mdpi.com/2076-3417/10/2/543data interpolationanomaly detectionbootstrapfine dustpm<sub>2.5</sub> |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
JinSoo Park Sungroul Kim |
spellingShingle |
JinSoo Park Sungroul Kim Improved Interpolation and Anomaly Detection for Personal PM<sub>2.5</sub> Measurement Applied Sciences data interpolation anomaly detection bootstrap fine dust pm<sub>2.5</sub> |
author_facet |
JinSoo Park Sungroul Kim |
author_sort |
JinSoo Park |
title |
Improved Interpolation and Anomaly Detection for Personal PM<sub>2.5</sub> Measurement |
title_short |
Improved Interpolation and Anomaly Detection for Personal PM<sub>2.5</sub> Measurement |
title_full |
Improved Interpolation and Anomaly Detection for Personal PM<sub>2.5</sub> Measurement |
title_fullStr |
Improved Interpolation and Anomaly Detection for Personal PM<sub>2.5</sub> Measurement |
title_full_unstemmed |
Improved Interpolation and Anomaly Detection for Personal PM<sub>2.5</sub> Measurement |
title_sort |
improved interpolation and anomaly detection for personal pm<sub>2.5</sub> measurement |
publisher |
MDPI AG |
series |
Applied Sciences |
issn |
2076-3417 |
publishDate |
2020-01-01 |
description |
With the development of technology, especially technologies related to artificial intelligence (AI), the fine-dust data acquired by various personal monitoring devices is of great value as training data for predicting future fine-dust concentrations and innovatively alerting people of potential danger. However, most of the fine-dust data obtained from those devices include either missing or abnormal data caused by various factors such as sensor malfunction, transmission errors, or storage errors. This paper presents methods to interpolate the missing data and detect anomalies in PM<sub>2.5</sub> time-series data. We validated the performance of our method by comparing ours to well-known existing methods using our personal PM<sub>2.5</sub> monitoring data. Our results showed that the proposed interpolation method achieves more than 25% improved results in root mean square error (RMSE) than do most existing methods, and the proposed anomaly detection method achieves fairly accurate results even for the case of the highly capricious fine-dust data. These proposed methods are expected to contribute greatly to improving the reliability of data. |
topic |
data interpolation anomaly detection bootstrap fine dust pm<sub>2.5</sub> |
url |
https://www.mdpi.com/2076-3417/10/2/543 |
work_keys_str_mv |
AT jinsoopark improvedinterpolationandanomalydetectionforpersonalpmsub25submeasurement AT sungroulkim improvedinterpolationandanomalydetectionforpersonalpmsub25submeasurement |
_version_ |
1725316752407003136 |