Decision Tree Model for Automatic Sterilizied Water Feeding System in Poultry Farm Based on Internet of Things

The automatic sterilized drinking water supply system for chickens is designed to facilitate the activities of farmers in the process of providing drinking water to chickens and to facilitate the process of sterilizing the water. The process of sterilizing water with UV light must be done properly in order to get good results. In this study, there were five HC-SR04 ultrasonic sensors connected to the ESP8266 microcontroller. The ESP8266 microcontroller is embedded with the decision tree method as an output decision maker based on the calculation of the C4.5 algorithm. There are three processes, namely the process of determining the dataset, forming a decision tree and forming a rule. From the results of several tests that have been carried out, it is known that the reading error values of each ultrasonic sensor are 3.63%, 2.09%, 1.54%, 3.39% and 3.38%, respectively. Then in testing the rules that have been generated from the decision tree method with a total of 10 test data, an accuracy of 100% is obtained. Average system execution time is 0.98 seconds.