Self-Adaptation of Software Services Based on User Profile

In recent years, several methods have been proposed to evaluate user experience (UX) improvements using real-time emotion detection, but the correlation between user emotions and personality traits in self-adaptive software has not been fully explored. This paper investigates the feasibility of implementing a software service that adapts based on the user's emotional and personality state. Using pre-trained models for real-time emotion detection and the BFI-2-S test for personality assessment, we conducted an experiment with 30 users on the Neo Digital Skills platform, evaluated through the User Experience Question-naire (UEQ). Our findings show that self-adaptation significantly enhances UX, with improvements in Attractiveness (from -0.81 to 1.69), Stimulation (from -0.73 to 1.78), and Novelty (from - 0.62 to 2.02), indicating that personalization based on emotions and personality traits boosts user engagement, satisfaction, and innovation perceptions.