Detect Nutrition on Food Recipes Using Machine Learning

Santoso, Muhammad Azka Aysar (2024) Detect Nutrition on Food Recipes Using Machine Learning. Diploma thesis, Institut Teknologi Sepuluh Nopember.

Text 5025201150-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2026. Download (7MB) | Request a copy

Abstract

Nutrition is substances related to health and disease, including all processes in the human body to receive food or materials from the environment and use these materials to carry out important activities in the body and excrete the rest. Nutrition serves to produce energy for organ function, movement and physical function, as a basic material for the formation and repair of body cell tissue and as a protector and regulator of body temperature. If a person experiences unbalanced nutrition, in the sense of lacking one or several nutrients, to experiencing overall malnutrition, it can cause impaired growth and health.
In this research, machine learning is used to analyse aspects of nutritional content based on food recipes. Determining attributes such as classification, taxonomy, and nutritional assessment of food is often a challenging and resource-intensive task, although it is essential for a better understanding of food. Evaluation is based on accuracy, precision, recall, and f1 score. The nutritional content in this food recipe is only limited to 8 nutrients, namely protein, carbohydrate, sodium, fibre, fat, potassium, cholesterol, and sugar. By using the right machine learning method, it is possible to obtain nutritional data from food recipes.
Based on the results of data testing analysis conducted with the Random Forest, Naive Bayes, and Decision Tree methods using the default hyperparameter technique and GridSearch, the results show that the best accuracy for detecting nutrients in a food recipe is the Random Forest method with an overall accuracy value of 0.97.

Item Type:	Thesis (Diploma)
Subjects:	T Technology > TD Environmental technology. Sanitary engineering > TD194.6 Environmental impact analysis
Divisions:	Faculty of Information and Communication Technology > Informatics > 55201-(S1) Undergraduate Thesis
Depositing User:	Muhammad Azka Aysar Santoso
Date Deposited:	01 Aug 2024 22:34
Last Modified:	01 Aug 2024 22:34
URI:	http://repository.its.ac.id/id/eprint/111648

Actions (login required)

View Item