Intelligent crop growth monitoring system using IoT and machine learning

Abstract

Agriculture is the important part of any developing country. As population increases the food requirement also be increases so the traditional farming not sufficient to fulfill the requirement of food. The development of industries, decreases in land space, insufficient rainfall, chemicals used in farming and peoples are moving towards the urban cities are the challenges of the agricultural field. So we have to use another source of food which we called as hydroponics. In hydroponics we can developed our plant by supplying the nutrients through water without use of soil. In this study we had used the applicability of machine learning algorithm like support vector regression(SVR), Linear Regression, Lasso Regression, Decision Tree(DT), Ridge regression and Random forest to find the crop growth accuracy from the result we found that random forest has given the highest accuracy 95% and the performance are evaluated on various model like Decision Tree Regression having better performance on (R2=0.86), Support Vector Regression having better performance on (MAE=12.65 and RMSE= 21.31) and Lasso Regression having better performance on (MSE=4.51). The whole dataset are stored on csv file and divided into two subpart 80% data is used for training and 20% is used for testing.