AI řízené plánování jídelníčku a analýza výživy využívající velké jazykové modely

Abstract

This bachelor's thesis focuses on the design and development of a mobile application that automates meal planning using large language models. The application is developed in React Native and generates personalized meal plans based on the user's profile, including recipes, nutritional values, and a shopping list. Communication with the OpenAI API is handled directly without a custom backend, and all user data is stored locally in a SQLite database. The thesis describes the complete model output processing pipeline, from prompt construction and response validation to ingredient normalization. The final part of the thesis is dedicated to the experimental evaluation of the application's reliability in meal plan generation and the assessment of user experience through a user survey.