.png){kind=icon}
.png){kind=icon}
Return
AI for enhanced water quality data imputation: a deep learning perspective
.png){kind=icon}
OA
AI
I
S
K
D
A
S
DOI：10.1039/d4va00367e.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Water quality data, a crucial resource for scientific water resource management practices (e.g., irrigation), engineering solutions (e.g., process control of both water and wastewater treatment plants), etc., are often hindered in their utility due to missingness within the dataset. Addressing this challenge, this perspective article underscores the necessity of missing data imputation. Along with highlighting the imputation strengths and limitations of different statistical and machine learning models, this article highlights deep learning (DL) models, and their underlying major limitations as well as potential resolutions. This study embodies novelty by proposing a robust model, integrating diverse solutions with an aim to set new standards in terms of accuracy, efficiency and adaptability in the domain of water quality data analysis. The paper presents the real-world implementation of the proposed framework along with its limitations and potential resolutions. Finally, the study concludes by calling forth coordinated efforts from researchers of diverse disciplines for developing a novel, generalized, and memory-efficient deep learning architecture.
Journal
E
IF:
4.4
Papers:
395
Citations:
1.0K
Organization
Cited Papers