Adaptive generative weighted recurrent neural network for dynamic media content recommendation: Leveraging continuous user feedback for real-time personalization

Abstract

Most effective ways to showcase your achievements in personalized marketing is employed for dynamic content and recommendations across websites, emails, and social media platforms. This type of content dynamically adapts with the help of user's behavior, preferences, and context, enhancing engagement through relevant and captivating experiences. Despite, decades of research that has greatly enhanced user experience through personalization, many challenges and complexities remain. To solve this research gap, a new dynamic content recommender is suggested for accurate user's modeling and preferences using generative artificial intelligence. The developed Adaptive Generative Weighted Recurrent Neural Network (AGWRNN)-based recommender system tackles the cold-start issues and provides appropriate content based on user feedback and preferences. The data required for providing the dynamic content recommendation is taken from public datasets. The proposed dynamic content recommendation model reduces information overload and increases user satisfaction. The AGWRNN model learns the review information following the current state of the user for providing content related to user preferences. The weight optimization takes place in AGWRNN using Enhanced Crayfish Optimization Algorithm (ECOA) for improving the dynamic content recommendation performance. The contextual patterns are effectively learned using this AGWRNN to generate new content for real-time personalization. The experimental evaluation is explored to validate the efficiency of the recommended dynamic content recommender system.