Implementacija priporočilnega sistema za hrano v aplikaciji Mapper

SIMČIČ, ANDRAŽ (author), Žabkar, Jure (mentor), Ispirova, Gordana (comentor)

V času vedno večjega ozaveščenja ljudi glede zdravja in prehranjevalnih navad je zelo težko priti do kakovostnih informacij o prehranskih vrednostih živil in obrokov. V ta namen na IJS razvijamo spletno aplikacijo Mapper in mobilno aplikacijo Eatvisor, ki bi pomagali pri reševanju te problematike. Spletna aplikacija Mapper je zasnovana kot orodje za nutricioniste, podjetja in trgovce za urejanje prehranskih hranilnih vrednosti živil in obrokov. Za prikaz, beleženje in urejanje prehranskih odločitev uporabnikov pa se razvija Eatvisor. Del vmesnika uporabniškega programa (API) omenjene aplikacije Mapper je tudi priporočilni sistem, ki je osrednja tema diplomske naloge. Da bi ta priporočilni sistem deloval že na samem začetku uporabe aplikacije, potrebujemo čim več ustreznih podatkov. Zato smo najprej ustrezno obdelali in uporabili informacije o živilih, obrokih, informacijah o udeležencih in drugih podatkih, pridobljenih v sklopu nacionalne raziskave SI.Menu 2017/18. Nato smo izdelali API, ki omogoča priporočanje hrane na podlagi pridobljenih podatkov in implementiranega algoritma. Za uporabo tega API, smo tudi nadgradili funkcionalnost aplikacije Eatvisor za prikaz priporočenih obrokov. Prispevek diplomske naloge je delujoč priporočilni sistem v mobilni aplikaciji Eatvisor, ki deluje preko nadgrajenega API-ja aplikacije Mapper.

• Language: Slovenian
• Keywords: priporočilni sistem, API, Mapper, OPKP, Eatvisor
• Work type: Bachelor thesis/paper
• Typology: 2.11 - Undergraduate Thesis
• Organization: FRI - Faculty of Computer and Information Science
• Year: 2023
• PID: 20.500.12556/RUL-144843
• COBISS.SI-ID: 148116739
• Publication date in RUL: 16.03.2023

Secondary language

• Language: English
• Title: Implementation of food recommendation system in Mapper application

Abstract

In an era of increasing awareness among people regarding health and dietary habits, it is very difficult to obtain quality information about the nutritional value of food and meals. To address this issue, we are developing the Mapper web application and the Eatvisor mobile application at IJS, which would help in solving this problem. The web application Mapper is designed as a tool for nutritionists, companies, and merchants to manage the nutritional values of foods and meals. Eatvisor is being developed for displaying, recording, and editing users' dietary decisions. One of the main themes of the thesis is the recommendation system, which is a part of the application programming interface (API) of the Mapper application. To ensure that the recommendation system works right from the start of using the application, we need as much relevant data as possible. Therefore, we first processed and used the information on food, meals, user information and other data obtained through the SI.Menu survey. We then developed an API that enables food recommendation based on the obtained data and the implemented algorithm. To use this API, we also upgraded the functionality of the Eatvisor application to display recommended meals. The contribution of the thesis is a functioning recommendation system in the Eatvisor mobile application, which operates through the upgraded API of the Mapper application.

• Keywords: recommendation system, API, Mapper, OPKP, Eatvisor