Soil Fertility Classification on Smart Agriculture GIS Application Using a Decision Tree Model

Smart Agriculture is an innovative solution to address the challenges of efficiency and sustainability in the agricultural sector, especially in remote areas. Integration of technologies such as Geographic Information System (GIS), Internet of Things (IoT), and cloud computing enables real-time monitoring of land conditions and supports data-based decision making. This study aims to design a soil fertility monitoring system that is easily accessible to farmers through digital devices. The proposed system uses a 7-in-1 soil sensor connected to ESP32 and Raspberry Pi to measure parameters such as nitrogen, phosphorus, potassium, pH, humidity, and temperature. The collected data is sent to a cloud-based database as input using the Decision Tree classification model in the application. The classification results are divided into three categories of soil fertility: Low, Medium, and High. Visualization of the results is done through a Flutter-based mobile application equipped with interactive GIS maps and API connectivity, allowing direct access to information for farmers. This system shows high performance with a model accuracy of ${9 9. 5 \%}$, while providing accurate, measurable, and user-friendly digital solutions to improve agricultural productivity and sustainability.