GIS와 공간 데이터마이닝을 이용한 교통사고의 공간적 패턴 분석 - 서울시 강남구를 사례로 -

A Study on Spatial Patterns of Traffic Accidents using GIS and Spatial Data Mining Methods: A Case Study of Kangnam-gu, Seoul

  • 발행 : 2004.06.01

초록

본 연구의 목적은 GIS와 공간 데이터마이닝 방법을 이용하여 교통사고의 공간적 패턴을 살펴보고 이웃한 공간 객체와의 공간적 연관성을 탐색하는 것이다. 이를 위하여 서울시 강남구 교통사고 데이터를 이용하여 공간적 경향 분석, 군집 분석 및 군집의 특성 기술, 이웃한 공간 객체와의 연관 분석을 실시하였다. 그 결과, 강남구의 교통사고는 특징적인 4개의 군집 유형을 통해 분류될 수 있으며, 각 군집별로 차별적인 특성들을 보여주고 있다. 또한, 교통사고의 발생 위치와 이웃한 공간 객체들과의 연관성에서는 공간 객체들의 개념수준이나 공간적 관계의 수준에 따라 다양한 규칙들이 발견되었다. 이러한 규칙들은 모두가 유의미하거나 흥미로울 수는 없지만, 맥락에 따라 다양하게 해석될 수 있으며, 보다 심화된 인구를 위한 새로운 가설들로 사용될 수 있을 것이다.

The purpose of this study is to analyze spatial patterns of traffic accidents and to investigate spatial relations among neighboring spatial objects by applying GIS and spatial data mining methods. This study investigated traffic accident data in Kangnam-gu, Seoul, as a case study. As a result, four clusters were emerged based on individual attributes of traffic accidents. Each cluster showed distinctive properties. In spatial associations between individual attributes of traffic accidents and neighboring spatial objects, there were many rules according to concept hierarchy and definition of spatial relations. Although all rules were not be interesting and significant, they could be a clue to investigate more.

키워드

참고문헌

  1. Advanced In Knowledge Discovery and Data Mining From data mining to knowledge discovery: an overview Fayyad,U.M.;Piatesky-Shapiro,G.;Smyth,P.;Fayyad,U.M.(ed.);Piatesky-Shapiro,G.(ed.);Smyth,P.(ed.);Ulthurusamy,R.(ed.)
  2. Intersection of Geospatial Information and Information Technology Data mining and knowledge discovery in the geographical domain, National Academic White Paper Gahegan,M.
  3. Advanced In Knowkedge Discovery and Data Mining Attribute-oriented induction in data mining Han,J.;Fu,Y.;Fayyad,U.M.(ed.);Piatesky-Shapiro,G.(ed.);Smyth,P.(ed.);Ulthurusamy,R.(ed.)
  4. IEEE Transactions on Knowledge and Data Engineering v.11 no.5 Discovery of multiple-level association rules from large databaes Han,J.;Fu,Y
  5. Data Mining:Concept and Techniques Han,J.;Kamber,M.
  6. Finding Groups in Data: an Introduction to Cluster Analysis Kaufman,L.;Rousseeuw,P.J.
  7. Proc. 4th Symposium on Large Spatial Databases(SSD'95) Discovery of spatial association rules in geographic information databases Koperski,K.;Han,J.
  8. Proc. International Symposium on Spatial Data Handing (SDH'98) An efficient two-step method for classification of spatial data Koperski,K.;Han,J.;Stefanovic,N.
  9. IEEE Transactions Knowledge in Data Engineering v.5 Systems for knowledge discovery in database Matheus,C.J.;Chan,P.K.;Piatetsky-Shapiro,G. https://doi.org/10.1109/69.250073
  10. ACM SIGKDD Exploration v.1 no.2 Discovering geographic Knowledge in data rich environ-ments: a report on a specialist meeting Miller,H.J.;Han,J. https://doi.org/10.1145/846183.846208
  11. Transactions of the Institute of British Geographer, New Series v.12 no.2 A Geography of road traffic accidents Whitelegg,J. https://doi.org/10.2307/622525