본문 바로가기
  • Home

Implementation of Image Based Café Atmosphere Classification and Recommendation System

  • Journal of The Korea Society of Computer and Information
  • Abbr : JKSCI
  • 2025, 30(4), pp.65~71
  • Publisher : The Korean Society Of Computer And Information
  • Research Area : Engineering > Computer Science
  • Received : February 25, 2025
  • Accepted : April 14, 2025
  • Published : April 30, 2025

Jongmyeon Jeong 1 Minseong Han 1 Seongmu Jo 1

1목포해양대학교

Accredited

ABSTRACT

In this paper, we implement an image-based café classification and recommendation system. Café images obtained through web crawling are classified using a pre-trained classifier based on the purpose or atmosphere of the café. Based on this classification, the system recommends nearby cafés that match the user’s preferred purpose or atmosphere. ResNet is employed for café image classification to achieve fast and accurate learning and inference. For the search component, ElasticSearch is adopted as the database engine to enable flexible search capabilities, including synonym matching and morpheme-based queries. The proposed method enhances the objectivity of recommendations by eliminating subjective bias from café owners or staff during the assignment of labels that describe the café’s atmosphere.

Citation status

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