PERSONALIZED HEALTH MONITORING DIET RECOMMENDER SYSTEM FOR ADULTS USING ANTHROPOMETRIC INDICES

Abstract

This thesis developed a personalized diet recommender system that enables users without knowledge about nutrition to get diet that best suits their needs based on their anthropometric measurements. The system is knowledge-based and receives information about his/her gender, age, activity level, height and weight to compute Body Mass Index (BMI), waist circumference and hip circumference to compute Waist to Hip Ratio (WHR). These basic information about an individual are combined using an adapted RETE pattern matching algorithm to classify the users. For each possible classification an existing meal plan generated with the assistance of a professional nutritionist is recommended. A set of rules are used to link the class to the meal plan from the user profile database to the diet database. At the usage of this system, the user gets to update the initial indices supplied and the system automatically computes the differences in the indices and helps the user monitor the progress made in his overall health status. Hypertext MarkUp Language (HTML), Cascading Style Sheet (CSS) and JavaScript were used for designing the front end, MySQL database as the back end and PHP as the scripting language. The database was implemented using AppServer. The system was evaluated by users which was done in terms of the user interface, availability of diets, accuracy, trustness, and satisfaction derived from the recommendations and the usage of the system. The overall rating gave an average result of 4.576 on a 5.0 scale. The result showed that the developed system was effective in recommending diets to the users and was easy to understand and use as it had a simple and beautiful user interface.