Machine Learning Model for Temperature and Humidity Automatic Control in Smart Poultry Farm

This study designed the Smart Poultry Farm system model using the Machine Learning KNN (K-Nearest Neighbor) method. The model building is based on the dataset obtained from simple if-then-else statements on real poultry. The iteration test was performed using various iteration numbers of 10, 100, 1,000, and 10,000 using 210 data sets. The K values tested are 1, 3, 5, 7, and 9. The most optimal accuracy value obtained from random samples is 97.36%; this value is obtained in research with two outputs, 10 iterations, and the Kvalue is 3.