APPLICATION OF MACHINE LEARNING WITH LINEAR REGRESSION MODEL TO ANALYSIS OF AGRICULTURAL COMMODITY PRODUCT QUALITY IN SUMEDANG REGENCY

Abstract

The agricultural sector is one of the main pillars of the economy in Sumedang Regency; however, the quality of yields from leading commodities such as rice, corn, and horticultural crops is still influenced by various complex and interrelated factors, including weather conditions, soil characteristics, and cultivation practices applied by farmers. This study aims to apply machine learning using a linear regression model to analyze the factors affecting the quality of agricultural commodity yields in Sumedang Regency. The method used is a quantitative method with an experimental research approach, as this study involves the development of a predictive model that is systematically tested and evaluated using numerical data, following the stages of the CRISP-DM framework or a general machine learning workflow, while also developing a predictive model that can serve as a decision-support tool for farmers and related stakeholders. The data used in this study include information on production, weather, soil quality, planting methods, fertilizer use, and rainfall, obtained from the local agricultural agency and field observations over a specific period. The research method consists of data collection, data preprocessing (cleaning and normalization), splitting the data into training and testing sets, and training the linear regression model using the Python programming language with the scikit-learn library. Model evaluation was conducted using Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and the coefficient of determination (R²) to measure the model's accuracy and predictive reliability. The results show that the linear regression model was able to identify significant variables affecting crop quality, such as rainfall, soil moisture, and fertilization intensity, with an adequate level of accuracy and an R² value indicating a strong relationship between the independent variables and harvest quality. The implementation of this model is expected to help farmers and stakeholders improve agricultural productivity sustainably, optimize resource allocation, and support data-driven agricultural policy-making in Sumedang Regency.