<![CDATA[This study discusses the application of Edge Intelligence in an Internet of Things (IoT)-based automatic plant watering system by utilizing the Decision Tree algorithm to support the concept of precision agriculture. The system was developed using several environmental parameters, such as soil moisture, air temperature, humidity, light intensity, and plant age as input data to determine watering decisions. All data processing is carried out directly on the ESP32 microcontroller so that the system can work faster, reduce dependence on cloud computing, and increase network and power usage efficiency. The Decision Tree method is applied through Information Gain calculations to determine the attributes that most influence watering decisions. Based on the analysis results, air humidity obtained the highest Information Gain value of 0.8813 bits at a threshold of 57.5%, making it considered the most effective in the watering classification process. However, soil moisture was chosen as the root node because it has a higher level of agronomic relevance and is easier to implement in the plant monitoring system. Data processing is carried out locally using the ESP32 microcontroller, while the monitoring data is sent via the MQTT protocol to a smartphone application for real-time monitoring. Test results show that the system is capable of carrying out automatic watering accurately, increasing water efficiency, and adapting to changing environmental conditions adaptively. Thus, this system can be an effective, intelligent, and sustainable solution in supporting IoT-based plant cultivation management. The results show that the Decision Tree algorithm can determine the right watering decisions through Information Gain calculations. The system is also connected to the MQTT protocol that allows real-time monitoring of plant conditions via a smartphone application. Based on the test results, the system is capable of carrying out automatic watering effectively, saving water, and adapting to changing environmental conditions. Therefore, this system can be an innovative and sustainable solution to support the development of IoT-based smart agriculture.]]>
Copyrights © 2026
