A Robust Regression Model Based on Optimal Feature Sets For Simple Decision Making In Indoor Farms

Abstract

This paper proposed a robust regression model for simple decision making in smart indoor farms by selecting optimal feature sets.  In our proposal, there are several steps to ensure the time-series data set which collected from sensor nodes in smart indoor farms are expanded to new features for regression analysis. Based on the feature sets, the useful data values which are high corelated with output predictor will be selected. The approach not only interpret curve fitting but also produces an optimal data set consisting of vector time series for simple decision making in smart indoor farms. Simulation results shown that our proposal can remove large number of independent variables while R-squared values of our proposed model and original model is almost same. Thus, a decision making can be given efficiently because out-come in our proposed model just focus on strong correlation of small independent variables.