Publication

Abstract : The “AI-Enhanced Product Recommendation System using YouTube Comments Analysis” leverages advanced deep learning models, specifically the Complex Convolutional Long Short-Term Memory (CLSTM) and Gated Recurrent Unit (CGRU). This system harnesses the rich and diverse user-generated content on YouTube, employing CGRU and CLSTM to analyze and extract valuable sentiments from comments associated with product videos. The CGRU model excels at capturing intricate sequential dependencies, while the CLSTM model excels at recognizing complex linguistic patterns. The synergy between these models enables the system to effectively decipher the subtle nuances and sentiments expressed in online discussions. The process involves collecting and preprocessing YouTube comments, training the models on sentiment analysis tasks, and utilizing the learned representations to enhance product recommendations. The system addresses the challenges posed by the informal and varied nature of user-generated content on YouTube, incorporating robust natural language processing techniques. The implementation of these advanced models contributes to the creation of a sophisticated recommendation system, offering insights derived from user interactions, ultimately empowering content creators, marketers, and platform moderators to make informed decisions in the dynamic digital landscape.