Analisis Ramalan Cuaca di Sekupang, Kota Batam Menggunakan Algoritma Decision Tree dan Confusion Matrix

Abstract

Weather is an atmospheric condition consisting of variables such as temperature, humidity, precipitation, wind, and air pressure and describes the state of the environment at a particular time and place. This research discusses the use of the Decision Tree algorithm in predicting conditions. Decision trees can model in the form of a tree structure, where each internal node represents a trait or feature, each branch represents a decision rule, and each leaf or leaflet represents a category or outcome. This research aims to predict the weather by applying data classification methods using the decision tree algorithm. The research method involves classifying weather data based on group attributes such as wind speed, temperature, rainfall, and weather. Decision tree is a model that maps decisions in the form of a tree. This research uses 104 historical weather data obtained from weather online, the dataset used includes time, temperature, cloud percent, rainfall, and weather. Test results with 70% of the dataset are useful for testing the model, and 30% are used to train model performance. Model evaluation is done using accuracy, precision, recall. The results show a model accuracy of 96.77%, indicating good model performance in predicting weather conditions.