A Comparative Study based on Random Forest and Support Vector Machine for Strawberry Production Forecasting

Abstract

The purpose of this research is to develop a prediction method for strawberry growth. The main focus is to assess and compare the performance of the three beds using random forest and support vector machine method, in which the correlation of strawberry growth is identified by using water quantity, site selection, soil preparation and parameters. In addition, the suggested study uses three types of beds that are compared to find out the best bed production of strawberry fruits. The dataset includes collected from greenhouse strawberry product in South Korea and the four predicted variables consist of Strawberry production Maebang, Sulhyang Production, Water Type A, Maebang, Sulhyang Production, Water Type B and Maebang, Sulhyang Production, Water Type C. Through three Rooms Strawberry production test, this research suggests that the factors would closely influence the strawberry growth in fruit concentration directly with the variation of harvesting by every day.