Exploring the Efforts of IAN-BGRU Justifications in Food Recommender System and its User Preferences

There has been a general increase in the frequency with which people discuss diet, eating, and cooking. In actuality, there are far more dishes to choose from and ingredient combinations to try, and chefs (and even IBM supercomputers) create new dishes all the time. Many people find it difficult to keep up with the ever-evolving landscape of accessible foods and eating habits, which contributes to the rising prevalence of chronic diseases like obesity and diabetes. Our recipe recommendation engine thus not only offers tasty but also nutritious options for the consumer. The proposed approach designed our software for mobile because the proposed system envisions it being utilized in many different rooms and environments, including kitchens. Data processing, feature selection, and model training are the first three stages of the suggested procedure. Normalization and encoding are two components of data processing. As a criterion for feature selection, they employ inconsistency. After features are selected, IAN-BGRU is used to train the models. The proposed technique outperforms IAN and BGRU, two of the most used alternatives.