Abstract: In contemporary society, numerous individuals face a wide array of health issues and concerns. Advising an appropriate diet is frequently difficult owing to the diverse needs, including weight loss, weight gain, and overall health maintenance, coupled with the constraints of time. To address this challenge, we embarked on developing a program aimed at promoting healthier eating habits. Our approach focuses on recommending only three categories of goods: those conducive to weight loss, weight gain, and maintaining general well-being. Our System of Dietary Recommendations relies on a comprehensive nutrient database, encompassing precise information about various nutrients. To tailor dietary suggestions, the system considers user inputs such as medical data and dietary preferences, including the choice between vegetarian and non-vegetarian meals within the aforementioned categories. In this discussion, we delve into the realms of food classification, essential parameters, and the application of machine learning techniques. The recommendation engine is built using Nearest Neighbours algorithm which is an unsupervised learner for implementing neighbour searches. It acts as a uniform interface to three different nearest neighbours algorithms: Ball Tree, KD Tree, and a brute-force algorithm based. For our case, utilizing the brute- force algorithm Cosine similarity is used due to its fast computation for small datasets.
Keywords: Diet Recommendation, Machine Learning, Clustering, Health Factors, Vegetarian and Non-vegetarian, Calories, BMI.
Cite:
Shreyas R, Shashank A, Pallavi A A, Likitha H T, Dr. Pallavi Barman,"Federated Learning Based Diet Recommendation System I", IJARCCE International Journal of Advanced Research in Computer and Communication Engineering, vol. 13, no. 3, 2024, Crossref https://doi.org/10.17148/IJARCCE.2024.133119.
|
DOI:
10.17148/IJARCCE.2024.133119
