Journal of Digital Contents Society (디지털콘텐츠학회 논문지)

Digital Contents Society (한국디지털콘텐츠학회)
권호 표지
DOI QR코드

DOI QR Code

A Study on Correlation Analysis and Preference Prediction for Point-of-Interest Recommendation

Point-of-Interest 추천을 위한 매장 간 상관관계 분석 및 선호도 예측 연구

  • Park, So-Hyun (Department of IT Engineering, Sookmyung Women's University) ;
  • Park, Young-Ho (Department of IT Engineering, Sookmyung Women's University) ;
  • Park, Eun-Young (Department of Visual Design, Hyupsung University) ;
  • Ihm, Sun-Young (Department of Big Data Research Center, Sookmyung Women's University)
  • 박소현 (숙명여자대학교 공과대학 IT공학과) ;
  • 박영호 (숙명여자대학교 공과대학 IT공학과) ;
  • 박은영 (협성대학교 시각디자인학과) ;
  • 임선영 (숙명여자대학교 빅데이터활용 연구센터)
  • Received : 2018.05.01
  • Accepted : 2018.05.25
  • Published : 2018.05.31
⟨ Previous Next ⟩

Abstract

Recently, the technology of recommendation of POI (Point of Interest) related technology is getting attention with the increase of big data related to consumers. Previous studies on POI recommendation systems have been limited to specific data sets. The problem is that if the study is carried out with this particular dataset, it may be suitable for the particular dataset. Therefore, this study analyzes the similarity and correlation between stores using the user visit data obtained from the integrated sensor installed in Seoul and Songjeong roads. Based on the results of the analysis, we study the preference prediction system which recommends the stores that new users are interested in. As a result of the experiment, various similarity and correlation analysis were carried out to obtain a list of relevant stores and a list of stores with low relevance. In addition, we performed a comparative experiment on the preference prediction accuracy under various conditions. As a result, it was confirmed that the jacquard similarity based item collaboration filtering method has higher accuracy than other methods.

최근 소비자관련 빅 데이터 증가와 함께 이와 관련된 기술인 POI(Point-of-Interest) 추천 기술이 주목받고 있다. POI란, 소비자가 흥미롭거나 유용하다고 여기는 특정한 장소를 의미한다. 이전에 진행되었던 POI 추천시스템 관련연구들은 특정 데이터 셋에 한정되어 과 적합 문제가 발생할 수 있다는 한계점이 있다. 따라서 본 연구에서는 서울로 및 송정로에 설치한 통합 센서로 부터 얻은 사용자 매장 방문 실 데이터를 이용하여 매장 간 유사도 및 상관관계를 분석하며, 분석 결과를 토대로 신규 사용자가 흥미 있을 만한 매장을 추천해 주는 선호도 예측 시스템 연구를 한다. 실험 결과, 다양한 유사도 및 상관관계 분석을 통하여 관련성이 높은 매장의 리스트와 관련성이 낮은 매장의 리스트를 도출해낼 수 있었다. 또한, 다양한 조건에서 선호도 예측 정확도를 비교 실험을 수행한 결과 자카드 유사도 기반 아이템 협업 필터링 기법이 타 방법에 비해 높은 정확도를 보이는 것을 확인할 수 있었다.

Keywords

Acknowledgement

Grant : (기반SW-창조씨앗2단계) SIAT형 CCTV 클라우드 플랫폼 기술 개발

Supported by : 정보통신기술진흥센터

References

  1. G. N. Lee, "Design of a Personalized POI Recommendation System in Location Based Social Networks", M.S. dissertation, Chungbuk National University, 2018.
  2. Y. H. Choi, S. Y. Lee, "Users' Moving Patterns Analysis for Personalized Product Recommendation in Offline Shopping Malls", Domestic Journal of Korean Institute of Intelligent Systems, Vol.16, No.2, pp. 185-190, 2006. https://doi.org/10.5391/JKIIS.2006.16.2.185
  3. D. H. Kim, J. M. Kim, S. W. Park, "Decision Tree Based Application Recommendation System," in Proceedings of Korean Institute of Information Scientists and Engineers, Vol.39, No.1(D), pp. 140-142, 2012.
  4. W. Luan, G. Liu, C. Jiang, "Partition-based Collaborative Tensor Factorization for POI Recommendation," International Journal of Automatica Sinica, Vol.4, No.3, pp.437-446, 2017. https://doi.org/10.1109/JAS.2017.7510538
  5. P. Zhao, X. Xu, Z. Zhou, K. Zheng, V. S. Sheng, H. Xiong, "Exploiting Hierarchical Structures for POI Recommendation", 2017 IEEE International Conference on Data Mining (ICDM), pp.655-664, 2017.
  6. M. J. Kim, S. J. Lee, "Measures of Abnormal User Activities in Online Comments Based on Cosine Similarity," Domestic Journal of The Korea Institute of Information Security and Cryptology, Vol.24, No.2, pp. 335-343, 2014. https://doi.org/10.13089/JKIISC.2014.24.2.335
  7. L. C. Han, "Quantitative Measure of the Changes of Migration Patterns Using Cosine Similarity," Domestic Journal of Korean Society of Rural Planning, Vol.23, No.2, pp. 67-74, 2017. https://doi.org/10.7851/Ksrp.2017.23.2.067
  8. S. H. Kim, J. G. Park, S. Y. Han, "A Framework to Evaluate Communication Quality of Operators in Nuclear Power Plants Using Cosine Similarity," Journal of The Korea Society of Computer and Information, Vol.15, No.9, pp. 165-172, 2010. https://doi.org/10.9708/jksci.2010.15.9.165
  9. Building recommendation engines : understand your data and user preferences to make intelligent, accurate, and profitable decisions, S. Gorakala, Packt Publishing
  10. R-Probability and Statistics, T. J. Lim, Life and Power Press
  11. Seoullo, http://seoullo7017.seoul.go.kr
  12. Naver maps, https://map.naver.com
  13. S. N. Lee, H. M. Kim, C. G. Lee, G. S. Lee, "Study on Automatic Bug Triage using Deep Learning)", Domestic Journal of Korean Institute of Information Scientists and Engineers, Vol.44, No.11, pp.1156-1164, 2017.

Cited by

  1. A Effective Method for Controlling Waiting Lines for Airport Immigration vol.20, pp.1, 2019, https://doi.org/10.9728/dcs.2019.20.1.89
  2. The Impact of Others' Choices on the Affective and Cognitive Processing When Choosing a Credit Card vol.20, pp.6, 2018, https://doi.org/10.9728/dcs.2019.20.6.1181
  3. User-friendly Korean conversation learning application using face swap and personalized recommendation vol.21, pp.12, 2018, https://doi.org/10.9728/dcs.2020.21.12.2125