Using Artificial Intelligence to Identify the Similarity of Architectural Styles: An Application Using Architecture Style Similarity Identifier

Abstract

This paper explores the use of Artificial Intelligence (AI) in identifying architectural style similarities. The research aims to maintain the beauty of the environment and traditional building forms around Borobudur Temple. It provides a critical review of previous important research that addresses the issue of using AI in architectural decision-making, particularly about architectural styles. The research resulted in an AI-assisted application called ASSI that can identify architectural style similarities. The application works by analysing building features and comparing them with a database of architectural styles using a CNN (Convolutional Neural Network) approach. The results show that the ASSI application can accurately identify architectural style similarities and can be used to assist architects and designers in creating new buildings that are in harmony with the traditional architecture of the region. Overall, this research makes a valuable contribution to the fields of architecture and AI and has practical applications for preserving traditional architecture in a historical context.