Hybrid Machine Learning Model for Electricity Consumption Prediction Using Random Forest and Artificial Neural Networks

Table 7

Comparison of predictor variables selected by SWR and RF feature selection methods.

Group of variables	SWR	RF 1	RF 2
Geospatial	Usage_Type; TOU	Usage_Type; Season	Usage_Type
Geographical	Population_Ratio	Area Population_Ratio	Area
Climatic	Mean_MSL_Pressure; Mean_Maximum_Temperature; Mean_Relative_Humidity; Total_Rainfall	Mean_Station_Pressure; Mean_Maximum_Temperature; Mean_Relative_Humidity; Mean_Minimum_Temperature; Mean_Drybulb; _Temperature; Mean_MSL_Pressure; Total_Rainfall	Mean_Station_Pressure; Mean_Maximum_Temperature; Mean_Relative_Humidity; Mean_Minimum_Temperature; Mean_Drybulb_ temperature
Industrial	Industrial_Labor; Industrial_Plant	Industrial_Labor; Industrial_Plant	Industrial_Labor
Household	Household; Liabilities	Household; Expenditure; Income; Agriculturists; Liabilities	Household; Expenditure; Income