Missing Value Imputation of Time-Series Air-Quality Data via Deep Neural Networks

International Journal of Environmental Research and Public Health

卷:18

Hunjoo Lee¹  Wonho Yang²  Jinhee Kim³  Jaegul Choo³  Taesung Kim³ 

[1] Department of Environmental Big Data, CHEM. I. NET, Ltd., Seoul 07964, Korea;

[2] Department of Occupation Health, Daegu Catholic University, Gyeongbuk 38430, Korea;

[3] Kim Jaechul Graduate School of Artificial Intelligence, KAIST, Daehak-ro 291, Yuseong-gu, Daejeon 34141, Korea;

关键词: time-series data;    spatio-temporal data;    missing value imputation;    interpretable deep learning;    air pollution;   

DOI  :  10.3390/ijerph182212213

来源: DOAJ

【 摘 要 】

To prevent severe air pollution, it is important to analyze time-series air quality data, but this is often challenging as the time-series data is usually partially missing, especially when it is collected from multiple locations simultaneously. To solve this problem, various deep-learning-based missing value imputation models have been proposed. However, often they are barely interpretable, which makes it difficult to analyze the imputed data. Thus, we propose a novel deep learning-based imputation model that achieves high interpretability as well as shows great performance in missing value imputation for spatio-temporal data. We verify the effectiveness of our method through quantitative and qualitative results on a publicly available air-quality dataset.

【 授权许可】

Unknown