The recommendation system utilizes deep learning techniques to learn hidden features of users and items from huge data and subsequently construct a recommendation model, finally generating effective recommendations for the user. Deep learning-based recommendation system outperforms conventional recommendation system due to their capabilities such as non-linear transformation, representation learning, sequence modeling, and flexibility. Most of the existing deep learning approaches for recommender systems focus on either user-s auxiliary information or item-s textual information. A new hybrid model approach with deep learning techniques aims to capture auxiliary information for user-s personal preferences and item-s textual information for different user-s interactions related to items. A hybrid deep learning-based recommendation system significantly enhances prediction accuracy and recommendation performance by exploring both user-s preferences and items attractiveness.