PENERAPAN MODEL REGRESI LINIER BERGANDA PADA SISTEM PREDIKSI PRODUKSI HASIL PERTANIAN(PADI) DI KOTA LUBUKLINGGAU DENGAN BAHASA PEMROGRAMAN PHYTON

Abstract

Lubuklinggau City is also one of the rice producing areas in South Sumatra province. Every year, rice production results in Lubuklinggau City experience unstable fluctuations up and down depending on the weather. In an effort to prevent price increases due to the uncertain amount of rice production, it is necessary to have precise and accurate predictions about harvest results in Lubuklinggau City. Prediction is the process of forecasting a variable in the future based more on intuitive considerations than on past data, although it places more emphasis on intuition. In price predictions, quantitative data is also often used as complementary information in forecasting. In the research carried out, the method used in the prediction system is the multiple linear regression method. The result of the research is the application of a multiple linear regression model to the agricultural production (rice) production prediction system in Lubuklinggau city using the Python programming language. From the test results, the Mean Absolute Error (MAE) value was 0.35609660506590457, the Mean Square Error (MSE) value was 0.174083367233848, the R2 value was 0.41723298914856677 and the Mean Absolute Percentage Error (MAPE) value was 22.179290114514938. From the results of the tests carried out it can be said that the resulting linear regression model is in the "Good" category. 

Keywords: Rice, Prediction, Multiple Linear Regression