DOI QR코드

DOI QR Code

Construction of Pilot System to Improve Search Quality in National Archives of Korea Portal and Effects Validation

국가기록포털 검색 품질 개선을 위한 파일럿 시스템 구축 및 실효성 검증

  • 소현기 (전북대학교 일반대학원 기록관리학과) ;
  • 염경록 ((주) 아이와즈) ;
  • 오효정 (전북대학교 문헌정보학과, 문화융복합아카이빙연구소)
  • Received : 2023.04.25
  • Accepted : 2023.05.19
  • Published : 2023.05.31

Abstract

The National Archives of Korea (NAK) operates the NAK Portal as a record search system. However, user search satisfaction is too low, and the number of visitors to the portal is gradually decreasing. This study identifies the portal's issues, proposes feasible improvements, and constructs a pilot system to validate the solutions. The preliminary assessment revealed six major issues, such as poor search tool performance and the lack of consistency in search results. After clarifying the improvement measures, a pilot system was established and compared with the National Records Portal. The evaluation showed significant performance improvements in the pilot system, such as Precision, Recall, and Mean Reciprocal Rank (MRR).

국가기록원에서는 국민에게 소장하고 있는 공공기록물에 대한 접근점을 제공하기 위해 국가기록포털이라는 대국민 검색서비스를 운영 중이다. 그러나 지속적으로 검색 결과에 대한 이용자 만족도가 낮다는 의견들이 수렴되고 있으며, 그와 더불어 포털 이용률도 감소하고 있는 추세이다. 본 연구는 이러한 상황을 극복하기 위해 수행한 국가기록포털 검색서비스 품질 점검 연구의 후속 연구로, 국가기록포털의 문제점을 규명하고 그에 따른 개선안을 제안하고, 나아가 그 실효성을 검증하는 것을 목표로 한다. 선행된 품질 평가를 통해 국가기록포털의 주요 문제점으로 검색 도구의 저조한 성능, 검색 결과의 일관성 결여, 기본검색 기능 부재를 비롯한 6가지 문제점을 도출하였으며 이에 대한 개선방안을 규명했다. 제안한 방안의 실효를 검증하기 위해 이 중에서 현실적으로 당장 도입 가능한 방안을 적용한 파일럿 시스템을 구축, 국가기록포털과 검색 성능 비교를 수행하였다. 평가 결과, 파일럿 시스템 검색 도구의 정확률, 재현율, MRR 모든 측면에서 유의미한 상승을 확인하였으며 그 효과를 입증하였다.

Keywords

Acknowledgement

본 논문은 년도 한국연구재단 연구비 지원에 의한 결과의 일부임(NRF-2021R1I1A3047435).

References

  1. Ahn, Se-Jin, Hwang, Hyeon-Ho & Lim, Jin-Hee (2022). A Case Study on the Application of AI-OCR for Data Transformation of Paper Records. Journal of the Korean Society for Information Management, 39(3), 165-193. https://doi.org/10.3743/KOSIM.2022.39.3.165
  2. Baek, Ji-Yeon & Oh, Hyo-Jung (2019). User Information Needs Analysis based on Query Log Big Data of the National Archives of Korea. Journal of the Korean Society for Information Management, 36(4), 183-205. https://doi.org/10.3743/KOSIM.2019.36.4.183
  3. Choi, Seon-Hee (2008). Improving the archival reference services through a use analysis of the national archives portal service. Master's thesis, Graduate School of Yonsei University, Library and Information Science.
  4. Jang, Hee-Jung (2012). A Study on Evaluation of National Archives Websites. Journal of Korean Society of Archives and Records Management, 12(2), 51-70. https://doi.org/10.14404/JKSARM.2012.12.2.051
  5. Jin, Joo-Young & Rieh, Hae-Young (2018). Analysis of Users' Inflow Route and Search Terms of the Korea National Archives' Web Site. Journal of the Korean Society for Information Management, 35(1), 183-203. https://doi.org/10.3743/KOSIM.2018.35.1.183
  6. Kang, Rye-Rim (2020) A Study on the Improvement of Archival Reference Service in the National Archives of Korea. Master's thesis, Graduate School of Ewha Womans University, Library and Information Science.
  7. Kang, Yoo-na, Jo, Young-jun, Kim, Min-Jung & Oh, Hyo-Jung (2022). Advancement Plans for Linkage of National Archives Portal Service to Improve Accessibility and Usability of National Records. Journal of the Korean Society for Information Management, 39(4), 99-125. https://doi.org/10.3743/KOSIM.2022.39.4.099
  8. Kim, Chul-Jin (2018). An Architecture and Development Process for Integration of Alexa-based Voice Service and RESTful API. Journal of Knowledge Information Technology and Systems, 13(3), 341-350. https://doi.org/10.34163/jkits.2018.13.3.005
  9. Kim, Ji-Hyun (2012). A Study on Users' Perception of Reference Services in National Archives of Korea. Journal of Korean Society of Archives and Records Management, 12(1), 167-187. https://doi.org/10.14404/JKSARM.2012.12.1.167
  10. Lee, Byung Gil & Kim, Heesop (2013). Design and Evaluation of an Individual Instance-based Ontology Retrieval System for Archival Records of the "Saemaul Movement". Journal of Korean Society of Archives and Records Management, 13(3), 67-97. https://doi.org/10.14404/JKSARM.2013.13.3.067
  11. Lee, Hyo-Jin & Kim, Ji-Hyeon (2021). A Study on Improvement Plans for National Archives of Korea Website's Search Service through Its Usability Evaluation. Journal of Korean Society of Archives and Records Management, 21(3), 187-215. https://doi.org/10.14404/JKSARM.2021.21.3.187
  12. Lee, Jong-Won, Jo, Woo-Seung & Kim, Tae-Hyun (2022). Design and Implementation of an Interactive Search System Based on KoBERT. Journal of Knowledge Information Technology and Systems, 17(5), 1,081-1,088. https://doi.org/10.34163/jkits.2022.17.5.028
  13. Lee, Yoon-Ryeong & Rieh, Hae-Young (2014). A Study on the Improvement Direction for Online Finding Aids: Based on the Assessment of National Archives. Journal of Korean Society of Archives and Records Management, 14(1), 75-100. https://doi.org/10.14404/JKSARM.2014.14.1.075
  14. Lee, Yu-Been & Rieh, Hae-Young (2017). A Suggestion of Interface for Ontology-Based Record Retrieval System. Journal of Korean Society of Archives and Records Management, 17(1), 217-244. https://doi.org/10.14404/JKSARM.2017.17.1.217
  15. Ministry of Science and ICT (2022). Survey on Internet use in 2021.
  16. Na, Jeong-Ho, So, Hyeon-Gi, Yeom, Gyung-Rok, Lee, Jung-Ok & Oh, Hyo-Jung (2022). Test Set Construction for Quality Evaluation of NAK Portal's Search Service and the Status Analysis. Journal of Korean Society of Archives and Records Management, 22(4), 25-43. https://doi.org/10.14404/JKSARM.2022.22.4.025
  17. Noh, Meung-Hoan (2020). The Way for Data Archive and Records/Archive Management in the 4th Industrial Revolution Era : Critical Reviews and Mid- and Long-term Proposals for the National Archives' Administrative Information Dataset Records/Archive Management Implementation Plan. The Korean Journal of Archival, Information and Cultural Studies, 11, 7-43. https://doi.org/10.23035/kaics.2020.1.11.007
  18. NAK (2015). White Paper on National Records, 2014. Daejeon: National Archives of Korea.
  19. NAK (2018). Plan to promote the organization of collection records in 2018. Daejeon: National Archives of Korea.
  20. NAK (2021a). Application and utilization of character recognition (OCR) technology of digitized records. Daejeon: National Archives of Korea.
  21. NAK (2021b). National Archives Major Statistical Yearbook 2021. Daejeon: National Archives of Korea.
  22. NAK (2022). Major business initiatives for 2022. Daejeon: National Archives of Korea.
  23. NAK (2023). Major business initiatives for 2023. Daejeon: National Archives of Korea.
  24. Park, Min-Su & Hyun, Mi-Hwan (2011). Usability of the National Science and Technology Information System. Journal of the Korean BIBLIA Society for Library and Information Science, 22(4), 5-19. https://doi.org/10.14699/KBIBLIA.2011.22.4.005
  25. Park, Sang-Hyun & Kim, Heesop (2014). Design and Implementation of an EAD-based Integrated Retrieval System for the Archives in Korea. Journal of Korean Society of Archives and Records Management, 14(1), 101-124. https://doi.org/10.14404/JKSARM.2014.14.1.101
  26. Park, Sun-Ho & Kim, Young-Kil (2019). Implementation of Disease Search System Based on Public Data using Open Source. Journal of the Korea Institute of Information and Communication Engineering, 23(11), 1,337-1,342. https://doi.org/10.6109/jkiice.2019.23.11.1337
  27. Seol, Moon-Won (2022). An Exploratory Investigation of Archival Reference Services in the National Archives of Korea. Journal of Korean Society of Archives and Records Management, 22(3), 103-124. https://doi.org/10.14404/JKSARM.2022.22.3.103
  28. Song, Ji-Sung & Jeong, Da-Hee (2021). A Study on the Usability of AI-based Naver App Search Service. Journal of the Korean Society of Design Culture, 27(1), 197-207. https://doi.org/10.18208/ksdc.2020.27.1.197
  29. Mihalcea, R. & Tarau, P. (2004). Textrank: Bringing order into text. In Proceedings of the 2004 conference on empirical methods in natural language processing, 404-411.
  30. Voorhees, E. M. (1999). TREC-8 Question Answering Track Report. Proceedings of the 8th Text Retrieval Conference, 77-82.