Penerapan Linear Sampling Dalam Komparasi Decision Tree Dan Deep Learning Pada Pelajar Pecandu Narkoba

Abstract

Drug addicts based on a report from the National Narcotics Agency (BNN) in Indonesia are more than 7 million people in the age range of 15-64 years. The number of students who become drug addicts is increasing every year, this phenomenon is very worrying because students are the next generation of the nation who are expected to become leaders in the future. In school areas, the pressure to excel and be accepted by peer groups often urges students to seek escape through the use of prohibited substances. Not only that, the lack of parental supervision and low understanding of the dangers of drugs also contribute to the increase in this addiction problem. This study compares the decision tree and deep learning methods to find the highest accuracy value between the two methods, while sampling is done using linear sampling. The highest accuracy value is in the decision tree method with an accuracy value of 69.50%, 100% precision and 0% recall.