XGBoost를 이용한 교통노드 및 교통링크 기반의 교통사고 예측모델 개발
Development of Traffic Accident Prediction Model Based on Traffic Node and Link Using XGBoost
2
조회
0
다운로드
1
인용
0
참고문헌
키워드
초록
This study intends to present a traffic node-based and link-based accident prediction models using XGBoost which is very excellent in performance among machine learning models, and to develop those models with sustainability and scalability. Also, we intend to present those models which predict the number of annual traffic accidents based on road types, weather conditions, and traffic information using XGBoost. To this end, data sets were constructed by collecting and preprocessing traffic accident information, road information, weather information, and traffic information. The SHAP method was used to identify the variables affecting the number of traffic accidents. The five main variables of the traffic node-based accident prediction model were snow cover, precipitation, the number of entering lanes and connected links, and slow speed. Otherwise, those of the traffic link-based accident prediction model were snow cover, precipitation, the number of lanes, road length, and slow speed. As the evaluation results of those models, the RMSE values of those models were each 0.2035 and 0.2107. In this study, only data from Sejong City were used to our models, but ours can be applied to all regions where traffic nodes and links are constructed. Therefore, our prediction models can be extended to a wider range.
Abstract
This study intends to present a traffic node-based and link-based accident prediction models using XGBoost which is very excellent in performance among machine learning models, and to develop those models with sustainability and scalability. Also, we intend to present those models which predict the number of annual traffic accidents based on road types, weather conditions, and traffic information using XGBoost. To this end, data sets were constructed by collecting and preprocessing traffic accident information, road information, weather information, and traffic information. The SHAP method was used to identify the variables affecting the number of traffic accidents. The five main variables of the traffic node-based accident prediction model were snow cover, precipitation, the number of entering lanes and connected links, and slow speed. Otherwise, those of the traffic link-based accident prediction model were snow cover, precipitation, the number of lanes, road length, and slow speed. As the evaluation results of those models, the RMSE values of those models were each 0.2035 and 0.2107. In this study, only data from Sejong City were used to our models, but ours can be applied to all regions where traffic nodes and links are constructed. Therefore, our prediction models can be extended to a wider range.
참고문헌 (18)
- Chen, T, "XGBoost: A Scalable Tree Boosting System", KDD’16, pp.785~794
- 조나혜, "도로 네트워크를 따른 교통사고 핫스팟의 시각화", 지적과 국토정보, 48(1), pp.201~213 (2018), 한국국토정보공사[DOI]
- Im, J.B, "Development of Accident Frequency Estimation Models Using Random Parameter", pp.2015
- 정용화, "판별모델 구축에 따른 도시고속도로의 교통사고 영향요인 분석에 관한 연구 - 부산지역 사례를 중심으로 -", 대한토목학회논문집(국문), 34(4), pp.1269~1278 (2014), 대한토목학회[DOI]
- Kang, Y.G, "Traffic Accident Frequency Prediction Model in Urban Signalized Intersections with Intelligent Theories", pp.2008
- 강영옥, "의사결정나무와 시공간 시각화를 통한 서울시 교통사고 심각도 요인 분석", 지적과 국토정보, 47(2), pp.233~254 (2017), 한국국토정보공사[DOI]
- "Korea Meteorological Agency Data Open Portal"[URL]
- Korea Ministry of Construction and Transportation, "Plan Intersection Design Guidelines" (2004)
- "Korea National Transport Information Center"[URL]
- "Korea Traffic Accident Analysis System"[URL]
- "Korean National Police Agency"[URL]
- 이경준, "도로위의 기상요인이 교통사고에 미치는 영향 - 부산지역을 중심으로 -", 한국데이터정보과학회지, 26(3), pp.661~668 (2015), 한국데이터정보과학회[DOI]
- 이상혁, "임의효과를 고려한 도심지 교차로 교통사고모형 개발에 관한 연구", 한국ITS학회 논문지, 14(1), pp.85~93 (2015), 한국ITS학회[DOI]
- Lundberg, S.M, "Advances in Neural Information Processing Systems 30" (2017)
- 오흥운, "도로선형설계요소의 표준편차를 이용한 설계일관성과 교통사고와의 상관성", 한국도로학회논문집, 11(2), pp.159~166 (2009), 한국도로학회
- 박민호, "확률적 모수를 고려한 음이항모형에 의한 교통사고와 기하구조와의 관계- 미국 워싱턴 주(州) 고속도로를 중심으로", 대한토목학회논문집(국문), 33(6), pp.2437~2445 (2013), 대한토목학회[DOI]
- Ryu, J.D, "Development of Expressway Traffic Accident Prediction Model Using Deep Learning", pp.2018
- "United States Geological Survey"[URL]