ENABLING NEXT-GENERATION SMART HOMES THROUGH AI-DRIVEN PERSONALIZED FOOD RECOMMENDATIONS

The rapid proliferation of advanced AI technologies has propelled numerous industries forward, but the smart home sector has yet to realize its full potential in the next-generation landscape. A true smart home transcends mere automation, becoming an entity that comprehends and anticipates residents' needs, providing timely and personalized services. This research paper explores the paradigm of a fully intelligent home environment, where residents enjoy a hospitality-like experience while their dwelling proactively caters to their requirements. One such provision involves offering customized food suggestions for daily meals, considering individual preferences, cultural influences, weather conditions, dietary restrictions, and an inclination to explore novel recipes. Our proposed system leverages a state-of-the-art NLP Bert model-based similarity prediction approach to rank recipes based on word and contextual similarities. Recipes sharing common ingredients and procedures are deemed similar, contributing to a diverse pool of options. By analyzing 'n' days of historical eating habits, the system generates a curated selection of the top 'k' recipes while avoiding repetition by excluding products consumed within the recent 'm' days (here m<<<<n).