A hybrid machine learning approach for enhanced anomaly detection in drinking water quality

ABSTRACT

Machine learning has become important for anomaly detection in water quality prediction. Data anomalies are often caused by the difficulties of analysing large amounts of data, both technical and human, but approaches have been inadequate when water contamination increases. The paper reports an anomaly detection technique based on an autoencoder and an extreme learning machine (AEELM). This approach uses an autoencoder for feature selection and an ELM algorithm for classification purposes. It is estimated using the Cauvery River dataset. The autoencoder-based ELM techniques outperformed other models with 83% accuracy, 77% precision, 83% recall, and 83% F1-level characteristic performance indicators. These results demonstrate its great effectiveness.

KEYWORDS:

• Anomaly detection
• autoencoder
• extreme machine learning

Disclosure statement

No potential conflict of interest was reported by the author(s).

Author contributions

J. Vellingiri has designed the algorithms, and Mr K. Kalaivanan has carried out the literature survey and performed implementation for our proposed system.

Data availability statement

Data are available based on request.

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.