Convolutional Neural Network-Based Personalized Program Recommendation System for Smart Television Users

The smart home culture is rapidly increasing across the globe and driving users toward utilizing appliances. Smart television (TV) one such appliance that embedded with technology. of TV have their interests in programs. However, automatic recommendation programs for user-to-user still under-researched. Several papers discussed systems, but those are related to different applications. Even though there some works on recommending (single-user multi-user), they did not discuss camera module capture validate user image personalized Hence, this paper proposes a convolutional neural network (CNN)-based program system users. To implement proposed approach, CNN algorithm trained datasets ‘CelebFaces Attribute Dataset’ ‘Labeled Faces Wild-People’ feature extraction detect human face. model applied captured by using module. Further, matched ‘synthetic dataset’. Based matching, hybrid filtering technique applied; thereby respective done. has achieved approximately 95% training performance. Besides, performance 85% from single-user perspective 81% multi-user perspective. From this, it observed outperformed conventional content-based collaborative techniques.