Klasifikasi Jenis Rumah Adat Malaka Menggunakan Metode Convulational Neural Network (CNN)

Abstract

In Indonesia, there is a rich diversity of cultures, one of which is traditional houses. Traditional houses essentially have the potential to represent the way of life, culture, and local economy. Traditional houses in Indonesia, particularly in the Malaka region, are important cultural symbols that can be regarded as cultural icons in Malaka and Indonesia. They provide a historical perspective, heritage, and reflect the progress of society in a civilization. The Convolutional Neural Network (CNN) method is used in this research. In this study, the CNN algorithm is applied to classify traditional house objects. These traditional house objects are divided into two categories: Kolibein Traditional House and Laleik Traditional House. The objective of this research is to classify traditional houses in Malaka, namely Kolibein Traditional House and Laleik Traditional House, and also to determine the accuracy level of CNN classification results. The previously created model is tested using test data to assess its accuracy. The testing is conducted on 20 data points, with 10 data points in each respective class. The testing results show that the classification of Kolibein and Laleik traditional houses is error- free or very accurate. Based on the model developed for classifying Kolibein and Laleik traditional houses using the Convolutional Neural Network method, it is evident that this method is capable of producing accurate results. The obtained results indicate that the accuracy, based on the classification report using images of Kolibein and Laleik traditional houses, reaches 100%. Therefore, it can be concluded that the constructed CNN model has a high level of accuracy.

Keywords

• classification method
• convolutional neural
• traditional house