Irrigation Prediction using Machine Learning in Precision Agriculture
Alifia Puspaningrum, A Sumarudin, Willly Permana Putra
Abstract
Agriculture has contributed to the economy in every country. Currently, agriculture has coped with several challenges, such as irrigation in water management. Agriculture used traditional techniques to manage irrigation. So, it requires much cost and labour. Besides, machine learning has evolved in various sectors to increase quality within. In a previous study, Fuzzy was proposed to model the data in statics based on rules and had difficulty computing the data automatically. This paper proposed irrigation prediction by using a machine learning algorithm. Classification algorithms, namely Support Vector Machine (SVM), K-Nearest Neighbour (KNN), Naive Bayes, Random Forest, and Decision Tree, are investigated to forecast accurate irrigation. This paper calculates accuracy, precision, recall, and Fl-score to evaluate the algorithm's performance. Experimental results show that Decision Tree outperforms other algorithms using its performance using the same agricultural data according to accuracy, recall, precision, and fl-measure score. In addition, the results of this experiment will be implemented in predicting automatic paddy irrigation using loT -based sensor data.