Association Rule for Product Delivery by Category wise

Personalized food recommendation engine suggests food to the user in a daily diet based on food orders pattern. Rule-based recommendation engine by category wise (Breakfast, Brunch, Lunch, Snacks, Dinner). The main objective of the rule, it should be more personalized and dynamically learn itself as data grow & customer feels user experience in the better way. Volume and Velocity of data are very limited. We have build multiple models based on data features / category.