Prediction Model of Indigofera Forage Production Using Regression Machine Learning Method

Forage sustainability is essential to support ruminants. Feed is the most significant contributor to the cost of raising livestock. Indigofera is the animal feeds that have high nuts, high nutrition, and production. This plant is an alternative forage that has been widely cultivated. Prediction of Indigofera forage production needs to be carried out so that the availability of forage is maintained. Factors that affect the amount of output are growth in the number of stems, stem length, many branches, and rainfall. The method used in this study was to use the machine learning method by building a Linear Regression method to predict Indigofera forage production. The data used was field observation data for three cycles based on rainfall in the Indigofera plant breeding area of UP3J Jonggol by dividing training data by 80% and test data by 20%. Data analysis using Python 3.19. This study employed an independent multivariable test on biomass production data. The coefficients of each independent variable are the number of stems 9.59, stem length 21.20, many branches -23.54, and rainfall 105.84. The accuracy level of the prediction model test is 91 %, so that the Indigofera biomass production model using the linear regression model is very good and can be applied.