Healthy eating plays a critical role in our daily well-being and is indispensable in preventing and managing conditions such as diabetes, high blood pressure, cancer, mental illnesses, asthma, and so on. In particular, for children and young people, the adoption of healthy dietary habits has been shown to be beneficial to early cognitive development. Healthy diets become more important in this COVID-19 pandemic situation to protect people by improving their immunity through a healthy food habit. Consequently, a healthful meal recommendation is getting much attention in the research community. In spite of these immense contributions, there are still a lack of successful systems that can model the arbitrary nature of meal recommendations. In particular, it becomes harder to recommend meals during special situations like illness. Because when a man or woman is sick their dietary habit or preference can be changed. Moreover, with the exponential increase in the number of available food options, it is not possible to take them all into account anymore. The only way to consider user preferences, maximize the number of healthy compounds and minimize the unhealthy ones in food, is using recommendation systems. To model these special cases, we are planning to leverage machine learning approach to build a recommendation system for foods: train, evaluate and test a model able to predict healthful foods upon user preferences and estimate the probability of negative food-drug interactions.