A Hybrid Movie Recommendation System Integrating Content-Based Filtering With Personality Traits
DOI:
https://doi.org/10.64252/xkhkh891Abstract
This paper presents a novel approach to movie recommendations by integrating traditional content-based filtering with personality traits analysis. We propose a hybrid system that combines TF-IDF-based similarity measures with the Big Five personality model to generate personalized movie recommendations. Our system analyzes movie features, including genres, cast, crew, and keywords, while incorporating user personality traits to adjust recommendations. Experimental results demonstrate improved recommendation diversity and user satisfaction compared to traditional content-based approaches. The hybrid approach achieves 12% higher precision and 15% better user satisfaction scores than content-based filtering alone, while maintaining computational efficiency. This work contributes to the growing field of psychologically informed recommendation systems and demonstrates the value of incorporating personality factors into content recommendation algorithms. Furthermore, our system provides insights into how personality influences movie preferences and enhances the overall recommendation experience. This work contributes to the growing field of psychologically informed recommendation systems and demonstrates the value of incorporating personality factors into content recommendation algorithms.