AI-Enhanced Emotional Prediction in Interior Design: Bridging Emotion and Architecture for Personalized Environments

Abstract

Artificial Intelligence can help in creating environments that adapt to users’ emotional states and personal preferences. This paper proposes an Artificial Intelligence-based framework to predict and personalize emotional experiences in interior spaces, addressing the limitations of subjective emotional architecture. The proposed framework integrates Kansei Engineering, fuzzy analytic hierarchy process (FAHP), and the Technique for Order Preference by Similarity to an Ideal Solution (TOPSIS) to translate users' emotional responses into weighted interior design parameters. These prioritized parameters are incorporated into an AI-based generative design model that produces personalized interior configurations. Emotional predictions are subsequently evaluated through immersive virtual reality (VR) environments by comparing AI-predicted emotional responses with physiological indicators and self-reported questionnaire data. The findings indicate that AI-assisted generative models effectively support participant-specific modifications to environmental design characteristics, including lighting, materials, color palettes, and biophilic integration. However, the framework demonstrated limited capability in generating coherent architectural floor plans, indicating that current AI systems are better suited to environmental personalization than comprehensive spatial planning.The research paper establishes a scalable model linking emotion prediction with adaptive, human-centered interior design to enhance well-being.