A Nutritional Recommender System for Rehabilitation of NCD’s by Using Data Mining Techniques

Abstract

Non Communicable or Chronic diseases tend to be long duration and are the results of a combination of genetic, phsycological, environmental and behaviour factors. About 41 million people died from NCD’s each year, which is equivalent to 70% of the global death toll. In many high-income countries, standard care for the long term management of NCD’s include Rehabilitation. So, in this paper I proposed that suitable Nutritional Diets play an important role in maintaining health, rehabitating the NCD’s and preventing the occurrence of NCD’s. For this purpose, I worked on a novel frame work named NutIngredientFood, which models the relationship between the Ingredients and their Proportions within food for the purpose of offering healthy recommendations.  Specifically, NutIngredientFood consists of three main components: 1) using an embedding based ingredient predictor to predict the relevant ingredients with user-defined initial ingredients(Ingredient Predictor), 2) predicting the amounts of the relevant ingredients(Amount Predictor), 3) creating a healthy pseudo-NutIngredient with a list of ingredients and their amounts according to the nutritional information and recommending the top similar nutritional ingredients with the pseudo-Nut Ingredient.