Journal of Intelligence and Information Systems (지능정보연구)

Korea Intelligent Information System Society (한국지능정보시스템학회)
권호 표지
DOI QR코드

DOI QR Code

Research Trends of Health Recommender Systems (HRS): Applying Citation Network Analysis and GraphSAGE

건강추천시스템(HRS) 연구 동향: 인용네트워크 분석과 GraphSAGE를 활용하여

  • Haryeom Jang (Department of Big Data Analytics, Kyung Hee Univeristy) ;
  • Jeesoo You (Department of Big Data Analytics, Kyung Hee Univeristy) ;
  • Sung-Byung Yang (Department of Business Administration/Big Data Analytics, Kyung Hee University)
  • 장하렴 (경희대학교 일반대학원 빅데이터응용학과) ;
  • 유지수 (경희대학교 일반대학원 빅데이터응용학과) ;
  • 양성병 (경희대학교 경영대학 경영학과 & 빅데이터응용학과)
  • Received : 2023.02.16
  • Accepted : 2023.03.05
  • Published : 2023.06.30
Download PDF
⟨ Previous Next ⟩

Abstract

With the development of information and communications technology (ICT) and big data technology, anyone can easily obtain and utilize vast amounts of data through the Internet. Therefore, the capability of selecting high-quality data from a large amount of information is becoming more important than the capability of just collecting them. This trend continues in academia; literature reviews, such as systematic and non-systematic reviews, have been conducted in various research fields to construct a healthy knowledge structure by selecting high-quality research from accumulated research materials. Meanwhile, after the COVID-19 pandemic, remote healthcare services, which have not been agreed upon, are allowed to a limited extent, and new healthcare services such as health recommender systems (HRS) equipped with artificial intelligence (AI) and big data technologies are in the spotlight. Although, in practice, HRS are considered one of the most important technologies to lead the future healthcare industry, literature review on HRS is relatively rare compared to other fields. In addition, although HRS are fields of convergence with a strong interdisciplinary nature, prior literature review studies have mainly applied either systematic or non-systematic review methods; hence, there are limitations in analyzing interactions or dynamic relationships with other research fields. Therefore, in this study, the overall network structure of HRS and surrounding research fields were identified using citation network analysis (CNA). Additionally, in this process, in order to address the problem that the latest papers are underestimated in their citation relationships, the GraphSAGE algorithm was applied. As a result, this study identified 'recommender system', 'wireless & IoT', 'computer vision', and 'text mining' as increasingly important research fields related to HRS research, and confirmed that 'personalization' and 'privacy' are emerging issues in HRS research. The study findings would provide both academic and practical insights into identifying the structure of the HRS research community, examining related research trends, and designing future HRS research directions.

현대사회는 정보통신기술 및 빅데이터 기술의 발전으로 누구나 인터넷을 통해 손쉽게 방대한 데이터를 얻고 활용할 수 있는 시대로, 양질의 데이터를 수집하는 능력을 넘어 수많은 정보 속에서 올바른 데이터만을 선별하는 능력이 더욱 중요해지고 있다. 이러한 기조는 학계에서도 이어지고 있는데, 축적되는 연구물 속에서 양질의 연구를 선별하여 올바른 지식구조를 형성하기 위해, 다양한 연구 분야에서 체계적 고찰(systematic review) 및 비체계적 고찰(non-systematic review)과 같은 문헌연구(literature review)가 수행되고 있다. 한편, 코로나19 팬데믹 이후 의료산업에서도 그동안 합의에 이르지 못했던 원격의료가 제한적으로나마 허용되고, 인공지능 및 빅데이터 기술이 응용된 건강추천시스템(health recommender systems: HRS)과 같은 새로운 의료서비스가 각광을 받고 있다. 하지만, 실무적으로 HRS가 미래 의료산업 발전을 이끌 중요한 기술로 평가받고 있음에도 불구하고, 학술적인 문헌연구는 다른 분야에 비해 매우 부족한 실정이다. 더불어 HRS는 학제적 성격이 강한 융합 분야임에도 불구하고, 기존의 문헌연구는 비체계적 고찰과 체계적 고찰 방법만을 주로 활용하여 이뤄졌기 때문에, 다른 연구 분야와의 상호작용이나 동적인 관계를 유추하기에는 한계가 존재한다. 이에, 본 연구에서는 인용네트워크 분석(citation network analysis: CNA)을 활용하여 HRS 및 주변 연구 분야의 전체적인 네트워크 구조를 파악하였다. 또한, 이 과정에서 최신 논문이 인용 관계가 잘 나타나지 않는 문제를 보완하기 위해 GraphSAGE 알고리즘을 적용함으로써, HRS 연구에 있어 'recommender system', 'wireless & IoT', 'computer vision', 'text mining' 등과 같은 연구 분야들의 중요도가 높아지고 있음을 파악하였으며, 이와 동시에 개인화(personalization) 및 개인정보보호(privacy) 등과 같은 새로운 키워드가 주요 이슈로 등장하고 있음을 확인하였다. 본 연구를 통해 HRS 연구 커뮤니티의 구조를 파악하고, 관련된 연구 동향을 살펴보며, 미래 HRS 연구 방향을 설계함에 있어 실질적인 통찰을 제공할 수 있을 것으로 기대한다.

Keywords

Acknowledgement

본 논문은 연구재단 4단계 BK21 사업으로부터 지원받은 연구임. 이 논문은 2022년 대한민국 정부(과학기술정보통신부)와 한국연구재단의 지원을 받아 연구되었음(NRF-2022K2A9A2A11097154).

References

  1. 강은경, 정연식, 양선욱, 권지윤, 양성병. (2022). MIS Quarterly 연구동향 탐색: 토픽모델링 및 키워드 네트워크 분석 활용. 지능정보연구, 28(2), 207-235. https://doi.org/10.13088/JIIS.2022.28.2.207
  2. 김기태. (2022). 한국과 유럽 8개국의 가구 의료비 지출 부담. 보건복지포럼, 18-33.
  3. 김현정, 안형식. (2015). 체계적 고찰/메타분석 연구 비판적으로 바라보기. The Korean Journal of Helicobacter and Upper Gastrointestinal Research, 15(2), 73-79. https://doi.org/10.7704/kjhugr.2015.15.2.73
  4. 김형철, 홍길종. (2022). 강원도 디지털 헬스케어 산업 경쟁력 분석과 시사점. The E-Business Studies, 23(5), 177-191.
  5. 문장섭. (2021). 정밀의료 시대의 뇌전증 관리. 에필리아: 뇌전증과사회, 3(2), 74-77.
  6. 박은희. (2021). 텍스트 네트워크 분석을 활용한 대학교양영어 연구주제 분석. Multimedia-Assisted Language Learning, 24(4), 214-239.
  7. 박준형, 곽기영. (2013). 특허 인용 관계가 기업성과에 미치는 영향. 지능정보연구, 19(3), 127-139. https://doi.org/10.13088/JIIS.2013.19.3.127
  8. 배영임, 신혜리. (2020). 코로나 19, 언택트 사회를 가속화하다. 이슈&진단, 1-26.
  9. 손진아. (2020). 제4차 산업혁명 시대에 있어서 문화산업의 발전조건. 한국자치행정학보, 34(1), 19-35. https://doi.org/10.18398/KJLGAS.2020.34.1.19
  10. 오의금. (2020). 포스트코로나 뉴노멀 시대의 간호의 전망과 대책. Korean Journal of Adult Nursing, 32(3), 221-222. https://doi.org/10.7475/kjan.2020.32.3.221
  11. 이동호, 김우수, 문성주, 정이상. (2012). 메타 데이터와 계량서지 데이터를 이용한 경영사학연구특성 분석. 경영사연구(경영사학), 27, 97-117.
  12. 이현복, 이수연. (2019). 노인의료비 추계와 건강보험에 대한 시사점. 보험학회지, 117, 43-68.
  13. 장요한. (2022). 목절별 통행량으로 보는 지역별 특색 읽기 1: 통계청 SDC 통신모바일 인구 이동량 통계데이터를 중심으로. 국토, 72-77.
  14. 정유경. (2020). 국내 문헌정보학 분야 학술지의 인용네트워크분석. 한국문헌정보학회지, 54(4), 221-238. https://doi.org/10.4275/KSLIS.2020.54.4.221
  15. 정이태, 안현철. (2020). 이종 네트워크 구조를 활용한 HinSAGE 기반 추천시스템. KMIS International Conference, 294-295.
  16. 정중희, 김종우. (2013). UCC 추천을 위한 사회 네트워크 분석을 활용한 협업 필터링 기법 연구. 한국정보기술학회논문지, 11(1), 185-195.
  17. 정혜실. (2014). 헬스케어 웨어러블 디바이스의 동향과 전망. KHIDI Brief, 115, 3-20.
  18. 조인동, 김남규. (2011). 소셜 네트워크와 데이터 마이닝기법을 활용한 학문 분야 중심 및 융합 키워드추천. 지능정보연구, 17(1), 127-138. https://doi.org/10.13088/JIIS.2011.17.1.127
  19. 채수미. (2020). 코로나바이러스감염증-19와 미래질병 대응을 위한 과제. 보건복지 Issue & Focus, 374, 1-8.
  20. 최건영, 홍참길. (2022). 약물 유인성 간염 예측을 위한 그래프 신경망 앙상블. 한국정보과학회 학술발표논문집, 1244-1246.
  21. Bassecoulard, E., & Zitt, M. (1999). Indicators in a research institute: A multi-level classification of scientific journals, Scientometrics, 44(3), 323-345. https://doi.org/10.1007/BF02458483
  22. Borgman, C. L., & Rice, R. E. (1992). The convergence of information science and communication: A bibliometric analysis, Journal of the American Society for Information Science, 43(6), 397-411. https://doi.org/10.1002/(SICI)1097-4571(199207)43:6<397::AID-ASI1>3.0.CO;2-M
  23. Calero Medina, C. M., & van Leeuwen, T. N. (2012). Seed journal citation network maps: A method based on network theory, Journal of the American Society for Information Science and Technology, 63(6), 1226-1234. https://doi.org/10.1002/asi.22631
  24. Calero Valdez, A., Ziefle, M., Verbert, K., Felfernig, A., & Holzinger, A. (2016). Recommender systems for health inormatics: State-of-the-art and future perspectives. Machine Learning for Health Informatics: State-of-the-Art and Future Challenges, 391-414.
  25. Cappella, J. N., Yang, S., & Lee, S. (2015). Constructing recommendation systems for effective health messages using content, collaborative, and hybrid algorithms. The ANNALS of the American Academy of Political and Social Science, 659(1), 290-306.
  26. Celik, E. D., & Elci, A. (2020). A survey on semanticized and personalized health recommender systems. Expert Systems, 37(4), e12519.
  27. Cheung, K. L., Durusu, D., Sui, X., & de Vries, H. (2019). How recommender systems could support and enhance computer-tailored digital health programs: A scoping review. Digital Health, 5, 1-19.
  28. Choi, J., Yi, S., & Lee, K. C. (2011). Analysis of keyword networks in MIS research and implications for predicting knowledge evolution. Information & Management, 48(8), 371-381. https://doi.org/10.1016/j.im.2011.09.004
  29. De Croon, R., Van Houdt, L., Htun, N. N., Stiglic, G., Vanden Abeele, V., & Verbert, K. (2021). Health recommender systems: Systematic review. Journal of Medical Internet Research, 23(6), e18035.
  30. Ding, Y., Chowdhury, G. G., & Foo, S. (2000). Journal as markers of intellectual space: Journal co-citation analysis of information retrieval area, 1987~1997, Scientometrics, 47(1), 55-73. https://doi.org/10.1023/A:1005665709109
  31. Etemadi, M., Abkenar, S. B., Ahmadzadeh, A., Kashani, M. H., Asghari, P., Akbari, M., & Mahdipour, E. (2023). A systematic review of healthcare recommender systems: Open issues, challenges, and techniques. Expert Systems With Applications, 231, 118823.
  32. Ferretto, L. R., Cervi, C. R., & de Marchi, A. C. B. (2017). Recommender systems in mobile apps for health a systematic review. In Proceedings of the 12th Iberian Conference on Information Systems and Technologies (CISTI), 1-6. IEEE.
  33. Goyal, P., & Ferrara, E. (2018). Graph embedding techniques, applications, and performance: A survey. Knowledge-Based Systems, 141, 78-94. https://doi.org/10.1016/j.knosys.2018.03.022
  34. Hamilton, W. L., Ying, R., & Leskovec, J. (2017). Inductive representation learning on large graphs. In Proceedings of the 31st Conference on Neural Information Processing Systems (NIPS 2017), 1-11.
  35. Hors-Fraile, S., Rivera-Romero, O., Schneider, F., Fernandez-Luque, L., Luna-Perejon, F., Civit-Balcells, A., & de Vries, H. (2018). Analyzing recommender systems for health promotion using a multidisciplinary taxonomy: A scoping review. International Journal of Medical Informatics, 114, 143-155. https://doi.org/10.1016/j.ijmedinf.2017.12.018
  36. Kamran, M., & Javed, A. (2015). A survey of recommender systems and their application in healthcare. Technical Journal (University of Engineering and Technology), 20(4), 111-119.
  37. Kim, J. A., Yoon, S., Kim, L. Y., & Kim, D. S. (2017). Towards actualizing the value potential of Korea Health Insurance Review and Assessment (HIRA) data as a resource for health research: Strengths, limitations, applications, and strategies for optimal use of HIRA data. Journal of Korean Medical Science, 32(5), 718-728. https://doi.org/10.3346/jkms.2017.32.5.718
  38. Lee, H. Y. (2015). Uncovering the multidisciplinary nature of technology management: Journal citation network analysis using the relatedness index, Scientometrics, 102(1), 51-75. https://doi.org/10.1007/s11192-014-1350-3
  39. Leppink, J., & Perez-Fuster, P. (2019). Social networks as an approach to systematic review. Health Professions Education, 5(3), 218-224.
  40. Leydesdorff, L. (1994). The generation of aggregated journal citation maps on the basis of the CDROM version of the Science Citation Index, Scientometrics, 31(1), 59-84. https://doi.org/10.1007/BF02018102
  41. Leydesdorff, L. (2007). Betweenness centrality as an indicator of the interdisciplinarity of scientific journals, Journal of the American Society for Information Science and Technology, 58(9), 1303-1319. https://doi.org/10.1002/asi.20614
  42. Liu, J., Ren, J., Zheng, W., Chi, L., Lee, I., & Xia, F. (2020). Web of scholars: A scholar knowledge graph. In Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval, 2153-2156.
  43. Mohamed, H. A., George, O. O., & Ronald, W. M. (2017). A survey of context-aware healthcare recommender systems. International Journal of Science and Research, 6(2), 962-967.
  44. National Center for Biotechnology Information (NCBI). (2023). PubMed. MEDLINE and PubMed Central, Feburary 3, 2023, from https://pubmed.ncbi.nlm.nih.gov/
  45. Noyons, E. C. M. (2001). Bibliometric mapping of science in a policy context, Scientometrics, 50(1), 83-98. https://doi.org/10.1023/A:1005694202977
  46. Otte, E., & Rousseau, R. (2002). Social network analysis: A powerful strategy, also for the information sciences, Journal of Information Science, 28(6), 441-453.
  47. Perozzi, B., AI-Rfou, R., & Skiena, S. (2014). Deepwalk: Online learning of social representations. In Proceedings of the 20th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 701-710.
  48. Pham, C., & Dang, T. (2021). Link prediction for biomedical network. In Proceedings of the 12th International Conference on Advances in Information Technology, 1-5.
  49. Pincay, J., Teran, L., & Portmann, E. (2019). Health recommender systems: A state-of-the-art review. In Proceedings of the 6th International Conference on eDemocracy & eGovernment (ICEDEG), 47-55. IEEE.
  50. Rafols, I., & Meyer, M. (2010). Diversity and network coherence as indicators of interdisciplinarity: Case studies in bionanoscience, Scientometrics, 82(2), 263-287. https://doi.org/10.1007/s11192-009-0041-y
  51. Ricci, F., Rokach, L., Shapira, B., & Kantor, P. B. (2011). Recommender systems Handbook. In F.Ricci, L. Rokach, B. Shapira, & P. B. Kantor. Recommender systems handbook. Springer US.
  52. Scarselli, F., Gori, M., Tsoi, A. C., Hagenbuchner, M., & Monfardini. G. (2009). The graph neural network model. IEEE Transactions on Neural Networks, 20(1), 61-80. https://doi.org/10.1109/TNN.2008.2005605
  53. Schafer, H., Hors-Fraile, S., Karumur, R. P., Calero Valdez, A., Said, A., Torkamaan, H., Ulmer, T., & Trattner, C. (2017). Towards health (AWARE) recommender systems. In Proceedings of the 2017 International Conference on Digital Health, 157-161.
  54. Scheuren, F. (2004). What is a survey. Alexandria: American Statistical Association.
  55. Sezgin, E., & Ozkan, S. (2013). A systematic literature review on Health Recommender Systems. In Proceedings of the 2013 E-Health and Bioengineering Conference (EHB), 1-4. IEEE.
  56. Sharma, D., Singh Aujla, G., & Bajaj, R. (2019). Evolution from ancient medication to human-centered Healthcare 4.0: A review on health care recommender systems. International Journal of Communication Systems, e4058.
  57. Sterling, J. A., & Montemore, M. M. (2022). Combining citation network information and text similarity for research article recommender systems. IEEE Access, 10, 16-23. https://doi.org/10.1109/ACCESS.2021.3137960
  58. Su, J., Guan, Y., Li, Y., Chen, W., Lv, H., & Yan, Y. (2020). Do recommender systems function in the health domain: A system review. arXiv e-prints, arXiv:2007.13058.
  59. Tran, T. N. T., Felfernig, A., Tranttner, C., & Holzinger, A. (2021). Recommender systems in the healthcare domain: State-of-the-art and research issues. Journal of Intelligent Information Systems, 57(1), 171-201. https://doi.org/10.1007/s10844-020-00633-6
  60. Wang, J. (2013). Citation time window choice for research impact evaluation. Scientometrics, 94(3), 851-872. https://doi.org/10.1007/s11192-012-0775-9
  61. Wang, H., Zhao, M., Xie, X., Li, W., & Guo, M. (2019). Knowledge graph convolutional networks for recommender systems. In Proceedings of the World Wide Web Conference, 3307-3313.
  62. White, H. D., & McCain, K. W. (1997). Visualization of literatures. Annual Review of Information Science and Technology, 32, 99-168.
  63. Wiesner, M., & Pfeifer, D. (2014). Health recommender systems: Concepts, requirements, technical basics and challenges. International Journal of Environmental Research and Public Health, 11(3), 2580-2607. https://doi.org/10.3390/ijerph110302580
  64. Wu, S., Sun, F., Zhang, W., Xie, X., & Cui, B. (2022). Graph neural networks in recommender systems: A survey. ACM Computing Surveys, 55(5), 1-37.
  65. Xiao, L., Wu, X., & Wang, G. (2019). Social network analysis based on graph. In Proceedings of the 12th International Symposium on Computational Intelligence and Design, ISCID 2019, 196-199.
  66. Yue, W., Wang, Z., Zhang, J., & Liu, X. (2021). An overview of recommendation techniques and their applications in healthcare. IEEE/CAA Journal of Automatica Sinica, 8(4), 701-717. https://doi.org/10.1109/JAS.2021.1003919
  67. Zhou, P., & Leydesdorff, L. (2007), The citation impacts and citation en-vironments of Chinese journals in mathematics, Scientometrics, 72(2), 185-200. https://doi.org/10.1007/s11192-007-1713-0