Klasifikasi Image Pokémon Menggunakan Convolutional Neural Networks

Abstract

This study aims to develop a Pokémon image classification model using Convolutional Neural Networks (CNN). The model is designed to identify the visual patterns of Pokémon based on their primary types, such as Water, Grass, Fire, Electric, and Poison. The research process includes dataset preparation using data sourced from Kaggle, image augmentation to enhance data variability, CNN model training, and model performance evaluation. The resulting model achieved an accuracy of 68.28% on the test data, indicating a reasonable ability to recognize complex visual patterns. Data augmentation and regularization techniques proved effective in improving the model’s generalization capability. This study is expected to contribute to the development of machine learning-based applications in digital entertainment, particularly in the recognition and classification of Pokémon characters. Future research is recommended to utilize larger datasets and integrate non-visual features to improve model performance further.