한국지리정보학회지 (Journal of the Korean Association of Geographic Information Studies)

  • 제26권1호
  • /
  • Pages.11-20
  • /
  • 2023
  • /
  • 1226-9719(pISSN)
  • /
  • 2287-6952(eISSN)
한국지리정보학회 (The Korean Association of Geographic Information Studies)
권호 표지
DOI QR코드

DOI QR Code

집단민원의 감성분석을 이용한 공간빅데이터 시각화 방안

A Study on the Visualization of Geospatial Big Data using Sentiment Analysis of Collective Civil Complaints

  • 주용진 (인하공업전문대학 공간정보빅데이터과)
  • Yong-Jin JOO (Dept. of Geospatial Big Data, Inha Technical College)
  • 투고 : 2022.12.11
  • 심사 : 2023.01.12
  • 발행 : 2023.03.31
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

전통적으로 공공 서비스에 대한 만족도 요인을 측정하기 위해 설문조사나 인터뷰 연구가 주를 이뤄 왔다. 민원의 단순 빈도를 떠나 민원에 내포된 감정의 경중까지 고려되지 않아 민원인이 체감하는 민원의 시급성, 고충의 심각 정도를 판단하기 어렵다. 이에 본 연구의 목적은 헤도노미터 단어별 행복도 점수를 활용해 집단민원이 내포하는 부정적 감성수치를 산정하는 방안을 제시하였다. 국민권익위원회의 2021년 지역별 상위 민원 토픽과 연관키워드 데이터를 대상으로 헤도노미터를 적용하여 민원의 주제별 부정적 감성수치를 산출하고, 지역별로 분포를 가시화하였다. 본 연구결과로 도출된 부정적 감성수치를 이용해 민원에 내포된 감정의 경중을 고려하여 민원인이 체감하는 민원의 시급성, 고충의 심각 정도를 판단하는데 도움이 될 수 있을 것으로 기대된다.

Traditionally, surveys or interview studies have been used to measure satisfaction factors for public services. This method focuses on the simple frequency of civil complaints and does not consider the aggravation of emotions implied in civil complaints. As a result, it is difficult to judge the urgency of civil complaints and the severity of grievances experienced by civil petitioners. This study aims to calculate the negative emotional value of collective complaints by using the happiness score for each word on the Hedonometer. The Anti-Corruption and Civil Rights Commission applied a Hedonometer to the top civil complaint topics and related keyword data by region in 2021 to calculate negative sentiment values by subject of civil complaints, and visualize the distribution by region. Using the negative emotional values derived from the results of this study, the severity of emotions contained in civil complaints can be considered. It is also expected to be helpful in determining the urgency of civil complaints and the severity of grievances experienced by civil petitioners.

키워드

과제정보

이 논문은 2022년도 인하공업전문대학 학술연구사업 지원에 의하여 연구되었음

참고문헌

  1. Daniel, P.P., G. Mihaela and A. Nikolaos, 2019, Automatically identifying complaints in social media, Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, Florence, Italy, July 28 - August 2, pp.5008-5019.
  2. Friendly, M. 2006. A brief history of data visualization. In: C. Chen et al.(ed.). Handbook of Data Visualization. Springer Handbooks Computational Statistics. Springer, Berlin, Heidelberg, pp.15-56.
  3. Friendly, M. 2009. Milestones in the history of thematic cartography, statistical graphics, and data visualization. http://www.math.usu.edu/~symanzik/teaching/2009_stat6560/Downloads/Friendly_milestone.pdf
  4. Hong, S.E. 2018. Sentimental & pattern analysis of environment complaint by big data mining. Inha University. Korea. 89pp
  5. Hwang, S.W. 2019. A study on the influence of happiness score on the movement of residence in the noise complaints: Focused on the complaint board in 25 districts of Seoul. Yonsei University. Korea. 75pp
  6. Jeon, S. E and D. B. Shin, 2018, A study on the agent based infection prediction model using space big data -focusing on MERS-CoV incident in Seoul. Journal of the Korean Association of Geographic Information Studies 21(2):94-106
  7. Jin, H. I., G.Y. Ock and M.s. Lee, 2020, A study on a methodology to measure happiness index with social big data: Focusing on Google Trends. Korean Journal of Korea society of innovation 15(5):215-237 https://doi.org/10.46251/INNOS.2020.12.15.5.215
  8. Kim, D.H. and D. Kim, 2018, Development and application of dynamic visualization model for spatial big data. Journal of the Korean Association of Geographic Information Studies 21(1):57-70
  9. Mikolov, T., W. T. Yih, and G. Zweig. 2018. Deconfounded lexicon induction for interpretable social science. In Proceedings of the 2018 Annual Conference of the North American Chapter of the Association for Computational, pp 1615-1625.
  10. Yang, W., L. Tan, C. Lu, A. Cui, H. Li, X. Chen, K. Xiong, M. Wang, M. Li, J. Pei, and J. Lin. 2019. Detecting customer complaint escalation with recurrent neural networks and manually-engineered features. In Proceedings of the 2019 Annual Conference of the North American Chapter of the Association for Computational Linguistics (Industry Track), NAACL, pp. 56-63.
  11. Zhou, G. and K. Ganesan. 2016. Linguistic understanding of complaints and praises in user reviews. In Proceedings of the 7th Workshop on Computational Approaches to Subjectivity, Sentiment and Social Media Analysis (WASSA), NAACL, pp. 109-114.