Improvement in Explicit Prediction of Water Quality Using Wavelet-Based LSSVR and M5pRT

Table 3

Various studies using different intelligent models.

Year	Author	Description	Parameters simulated	Results limited to
2014	Akrami SA. et al.	Rainfall data analyzing using moving average (MA) model and wavelet multiresolution intelligent model for noise evaluation to improve the forecasting accuracy	Wavelet transform (WT), moving average (MA)	RMSE and R^2 computed for MA at various levels of WT.
2018	Mahmoodabadi M. and Rezaei Arshad R.	Evaluated water quality parameters of the Karoun River using a regression approach and adaptive neuro-fuzzy inference system	Mann–Kendall regression, ANFIS	RMSE, MAE, and R^2 computed for the ANFIS model.
2019	Salazar L. et al.	Hourly ozone concentrations predicted using wavelets and ARIMA models	Haar discrete wavelet transform (HDWT), ARIMA	MSE and MSPE computed for ARIMA, HDWT, and combine model. The combined model performed better.
2019	Dehghani M. et al.	Predicted hydropower generation using the grey wolf optimization adaptive neuro-fuzzy inference system	ANFIS and GWO-ANFIS	RMSE, MAE, R^2, relative error, and confidence index computed for ANFIS and GWO-ANFIS. GWO-ANFIS was observed to be better.
2020	Seifi A. and Riahi H.	Estimated daily reference evapotranspiration using hybrid gamma test-least square support vector machine, gamma test-ANN, and gamma test-ANFIS models in an arid area of Iran	LSSVM, ANN, and ANFIS (all with gamma parameter)	RMSE, MAE, and R^2 computed for LSSVR, ANN, and ANFIS. LSSVR performed well overall.
2020	Present study Bhardwaj R., Bangia A.	Improved explicit prediction of river water quality using wavelet- based LSSVR and M5pRT	LSSVR, M5pRT, WLSSVR, WM5pRT	MSE, RMSE, MAE, and R^2 computed for LSSVR, M5pRT, WLSSVR, and WM5pRT. Wavelet conjuncted LSSVR and M5pRT observed to be better prediction models in our study.