Movie Recommender System Based on Generative AI

Movie recommender systems have revolutionized how people discover and experience movies using sophisticated algorithms, such a system generates personalized recommendations to assist users in reaching the movies that precisely fit their tastes. The traditional movie recommender systems lacks of diversity, unable to capture complex user preferences, and suffers from the cold-start issues. In this paper, we utilize the generative AI, specifically with Variational Autoencoders (VAE), to solve these problems. By combining VAEs, the system aims to produce more rich and individualized recommendations. VAE mitigates these issues by enabling the generation of meaningful item representations, promoting diversity in recommendations, and capturing the complexity of user preferences in a continuous latent space. The implementation is structured as follows: first, review the generative AI in movie recommender systems; second, implement the chosen generative AI approach; and finally, conduct the performance metrics evaluation of the selected generative AI approach. Ultimately, this research will enable the system to offer an achievement of an interactive, changing, and more personalized movie-discovering experience through the amazing power of generative AI.