Title: A machine learning-based food recommendation system with nutrition estimation

Authors: Anupama Nandeppanavar; Medha Kudari; Prasanna Bammigatti; Kaveri Vakkund

Addresses: Department of Master of Computer Applications, KLE Institute of Technology, Opp. Airport, Gokul, Hubballi, 580027, India ' Department of Master of Computer Applications, KLE Institute of Technology, Opp. Airport, Gokul, Hubballi, 580027, India ' Department of Computer Science and Engineering, KLE Institute of Technology, Opp. Airport, Gokul, Hubballi, 580027, India ' Department of Master of Computer Applications, KLE Institute of Technology, Opp. Airport, Gokul, Hubballi, 580027, India

Abstract: The human body needs energy to perform various activities, which are provided by calories. The proposed work is an efficient, user-friendly tool to assist calorie calculation. The system takes inputs such as height, weight, age, gender, and daily exercise level to estimate the recommended daily caloric intake. To achieve this, three machine learning models, K-nearest neighbours (KNN), decision tree and random forest algorithms, are employed to enhance the accuracy of predictions. Model accuracy achieved is 96.4% for KNN, 97.1% for decision tree and 96.8% using random forest algorithms. In addition to providing personalised caloric intake recommendations, the proposed system also offers diet recipes for breakfast, lunch and dinner tailored to the individuals's specific needs and preferences. Through the integration of machine learning algorithms, a user-friendly GUI, and personalised diet recommendations, the project aims to promote healthier eating habits and overall well-being for users.

Keywords: accuracy; BMI; body mass index; calorie; data processing; dietary; recipes; user interface visualisation.

DOI: 10.1504/IJDATS.2024.142485

International Journal of Data Analysis Techniques and Strategies, 2024 Vol.16 No.4, pp.487 - 507

Received: 07 Nov 2023
Accepted: 08 Apr 2024
Published online: 04 Nov 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article