Table 9.

A summary of AI-Big data analtiycs models for water monitoring

Ref.	AI model	Task	Building nature	Year	Description	Evaluation metrics					Others
						RMSE	MAE	MAPE	ACC	F1
Altunkaynak and Nigussie (2017)	MLP	Water demand prediction			Predict water demand using MSA-MLP and compare its DWT-MLP	$✓$					CE
Zubaidi et al. (2018)	GSA-ANN, BSA-ANN	Water demand prediction	Residential	2018	Predict water demand using heuristis algorithms, ANN and weather variables	$✓$
Chen et al. (2020)	DT, NB, LR, LDA,CRT, KNN, SVM, RF, CRF	WQP	HEP plant	2020	Predict water quality using different water paranmetersi.e. pH, DO, CODMn, and NH3–N					$✓$	wF1
Shine et al. (2018)	RF, NNN, SVMCDT	Water consumption prediction	Agricultural	2018	Predict water consumption using a backward sequential variable selection and parameter tuning			$✓$	$✓$
Smolak et al. (2020)	RF, SVM, ARIMA	Water consumption prediction	Residential	2020	Predict water consumption using consumption records and occupancy patterns	$✓$	$✓$	$✓$	$✓$
Antunes et al. (2018)	ANN, RF, SVM KNN	Water demand prediction	Public	2018	Reliable prediction while no significant anomalies of the data used during training are reported	$✓$		$✓$			$R^2$
Roccetti et al. (2019)	RNN	Predicting water meter failures	-	2019	Predict water meter failures using 15 million of readings				$✓$	$✓$	AUC, CM
Nasser et al. (2020)	LSTM	Water demand prediction	Public and residential	2020	Predict energy demand by analyzing data gathered from smart IoT water meters and stored in the cloud	$✓$	$✓$	$✓$
Du et al. (2021)	LSTM-DWT-PCA	Water demand prediction	Public and residential		The outputs of DWT and PCA are fed into an LSTM network to predict water demand	$✓$		$✓$			EVS, $R^2$