Perbandingan Metode KNN, Naive Bayes, dan Regresi Logistik Binomial dalam Pengklasifikasian Status Ekonomi Negara

Abstract

The classification of a country's economic status as developed or developing often involves factors such as life expectancy and its underlying variables. This research aims to compare the performance of three machine learning algorithms, namely KNN (K-Nearest Neighbors), naive Bayes, and binomial logistic regression, in classifying the economic status of countries as developed or developing. The data used in this study is "Life Expectancy (WHO) Fixed," obtained from the Kaggle website. The first statistical analysis conducted was Principal Component Analysis (PCA) using 16 predictor variables. PCA resulted in three principal components capable of explaining 71.41% of the variance, which were subsequently used in the KNN, naive Bayes, and binomial logistic regression methods. The analysis results from the KNN, naive Bayes, and binomial logistic regression methods produced F1-scores of 100\%, 98.19%, and 97.36%, respectively.

Comparison of Naive Bayes, K-Nearest Neighbor, and Support Vector Machine Classification Methods in Semi-Supervised Learning for Sentiment Analysis of Kereta Cepat Jakarta Bandung (KCJB)

“…Naïve Bayes is a classification method that requires calculating an opportunity for each attribute to build a Naïve Bayes model. This method requires instructions to determine the class for sample data [21]. • K-NN.…”

“…Non-parametric algorithms are computationally slower but make fewer assumptions about the data [23]. This algorithm works by classifying data into several classes from several data as the nearest neighbors of that point [21]. • SVM.…”