Analisis Klasifikasi Popularitas Game Roblox Menggunakan Algoritma K-Nearest Neighbor (KNN)

Abstract

Analysis of Roblox Game Popularity Classification Using the K-Nearest Neighbor (KNN) Algorithm is conducted to develop a data-driven classification model capable of mapping popularity levels of games on the Roblox platform. The dataset consists of numerical attributes such as total visits, active players, and rating ratios, which present heterogeneous characteristics and thus require a non-parametric approach. The K-Nearest Neighbor (K-NN) algorithm is applied due to its ability to measure similarity through Euclidean distance without assuming any specific data distribution. The research procedure includes data preprocessing, normalization, feature extraction, selection of k values, and mapping of popularity classes. Model performance is evaluated using accuracy, precision, recall, and F1-score. The results indicate that K-NN achieves stable performance across different k values and effectively distinguishes Roblox game popularity categories. These findings confirm that K-NN is a simple yet accurate method for data-driven game popularity analysis.