본문 바로가기
  • Home

Design of a Personalized Food Recommender Using GPS, TF-IDF, and Reinforcement Learning

  • Journal of The Korea Society of Computer and Information
  • Abbr : JKSCI
  • 2025, 30(6), pp.91~99
  • Publisher : The Korean Society Of Computer And Information
  • Research Area : Engineering > Computer Science
  • Received : May 7, 2025
  • Accepted : June 12, 2025
  • Published : June 30, 2025

Sang Kyoung Shin 1 Min Soo Sun 1 Min Young Kim 1 Ho-Young Kwak 1 Soo Kyun Kim 1

1제주대학교

Accredited

ABSTRACT

In modern society's busy daily life and information overload, many people struggle with excessive time and energy consumption even when choosing lunch menus due to numerous options, ultimately leading to decision fatigue and decreased satisfaction. Particularly, office workers and students need quick decisions during limited lunch breaks, yet effective decision-making tools are lacking. To address this issue, we design "WhatToEat?", a personalized menu recommendation service utilizing GPS and user preferences. We implement a mobile application based on React Native and Django frameworks. By optimizing the food menu recommendation system using TF-IDF algorithms and enhancing nutritional information features for allergies and dietary adjustments through reinforcement learning, we improved user satisfaction by up to 60%.

Citation status

* References for papers published after 2023 are currently being built.

This paper was written with support from the National Research Foundation of Korea.