Automatic vs. manual feature engineering for anomaly detection of drinking-water quality
Abstract
This paper evaluates anomaly detection approaches for drinking-water quality. Two major machine learning techniques are compared. One is manual feature engineering with feature subset selection for dimensionality reduction. The other is automatic feature learning through a recurrent neural network. Both methods incorporate the time domain for change detection. Preliminary results show a superior performance of automatic feature learning with an F1 score of 80%. While the feature set proposed in this work out-performs naive classification with original features, it needs further analysis to reach comparable performance to the automatic approach.
Read the paper
Related studies
- Green Chemistry: A Framework for a Sustainable Future
- Nano/micro-plastic, an invisible threat getting into the brain
- Emerging contaminants antibiotic resistance genes and microplastics in the environment: Introduction to 21 review articles published in CREST during 2018–2022
- Water Resources and Sustainable Development
- Nano- and microplastics in the workplace
- Water Enclosure and World-Literature: New Perspectives on Hydro-Power and World-Ecology
Go deeper in Evidence Hub
Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.