Loading...
Smart Resturantordering System with AI-Powered Personliztion and secure Real-Time Management
Hussaini, Farha Hidhaa, Lama Al Aulaqi, Dana
Hussaini, Farha
Hidhaa, Lama
Al Aulaqi, Dana
Citations
Altmetric:
Type
Supervisor
Date
Collections
Research Projects
Organizational Units
Journal Issue
Abstract
This paper presents the design, implementation, and evaluation of a hybrid recommendation system aimed at enhancing personalized customer experiences within smart restaurant ordering environments. The proposed model integrates multiple recommendation strategies—collaborative filtering, content-based filtering, and popularity-based filtering—into a unified architecture using a top meta-learner hybrid approach. By combining these techniques, the system is able to capture both user-specific behavioral patterns and item-specific attributes, while also considering overall content popularity to boost recommendation relevance. Customer order histories and detailed item metadata (such as cuisine type, dietary tags, and taste profiles) were used to train and validate the system. The hybrid model dynamically learns how to weigh each recommendation strategy through a meta-learning framework, optimizing prediction accuracy based on contextual relevance and user preferences. The system’s performance was rigorously evaluated using standard metrics, including precision, recall, F1-score, and hit rate, across various top-N recommendation thresholds. Our analysis reveals that the hybrid approach significantly outperforms traditional single-method recommendation systems in aligning suggestions with actual customer behavior, particularly in scenarios requiring personalization and adaptability. Furthermore, the study explores how varying the number of recommended items impacts system performance, offering valuable insights into balancing breadth and relevance in recommendations. Limitations of the current model—such as scalability and cold-start challenges—are discussed, along with potential enhancements and directions for future research. This work contributes to the field of intelligent recommendation systems by demonstrating the effectiveness of a meta-learner-based hybrid framework in real-world applications and offering a scalable solution for personalized, context-aware suggestions in the food service industry.