Browse > Article
http://dx.doi.org/10.13161/kibim.2022.12.2.001

Machine Learning based Optimal Location Modeling for Children's Smart Pedestrian Crosswalk: A Case Study of Changwon-si  

Lee, Suhyeon (한국외국어대학교 경영학과)
Suh, Youngwon (한국외국어대학교 경영학과)
Kim, Sein (한국외국어대학교 경영학과)
Lee, Jaekyung (홍익대학교 도시공학과)
Yun, Wonjoo (한국외국어대학교 경영학과)
Publication Information
Journal of KIBIM / v.12, no.2, 2022 , pp. 1-11 More about this Journal
Abstract
Road traffic accidents (RTAs) are the leading cause of accidental death among children. RTA reduction is becoming an increasingly important social issue among children. Municipalities aim to resolve this issue by introducing "Smart Pedestrian Crosswalks" that help prevent traffic accidents near children's facilities. Nonetheless such facilities tend to be installed in relatively limited number of areas, such as the school zone. In order for budget allocation to be efficient and policy effects maximized, optimal location selection based on machine learning is needed. In this paper, we employ machine learning models to select the optimal locations for smart pedestrian crosswalks to reduce the RTAs of children. This study develops an optimal location index using variable importance measures. By using k-means clustering method, the authors classified the crosswalks into three types after the optimal location selection. This study has broadened the scope of research in relation to smart crosswalks and traffic safety. Also, the study serves as a unique contribution by integrating policy design decisions based on public and open data.
Keywords
Smart Pedestrian Crosswalk; Optimal Location; Big Data; Machine Learning; Geographic Information System(GIS);
Citations & Related Records
Times Cited By KSCI : 2  (Citation Analysis)
연도 인용수 순위
1 An, K. E., Lee, S. W., Jeong, Y. J., Lee, M. W., Lee, D. H., Seo, D. M. (2016). A Study of Object Detection in Crosswalk Area using Smart Crosswalk System, Proceedings of the Korean Information Science Society Conference, pp. 1,498-1,500.
2 BeaK, T. H., Park, B. H. (2013). Developing Children Traffic Accident Models in School Zone : Focused on Cheongju, Proceedings of the KOR-KST Conference, 68, pp. 134-138.
3 Oh, S. H., Baek, J. H., Kang, U. G. (2021). Classification models for chemotherapy recommendation using LGBM for the patients with colorectal cancer, Journal of The Korea Society of Computer and Information, 26(7), pp. 9-17.   DOI
4 Park, S. N., Lim, J. B., Kim, H. K., Lee, S. B. (2017). Accidents involving Children in School Zones Study to identify the key influencing factors, International Journal of Highway Engineering, 19(2), pp. 167-174.   DOI
5 Rousseeuw, P. J. (1987). Silhouettes: a graphical aid to the interpretation and validation of cluster analysis, Journal of computational and applied mathematics, 20, pp. 53-65.   DOI
6 TAAS. (2020), 2019 Traffic Accident Statistical Analysis Report, http://taas.koroad.or.kr/ (Dec. 17. 2021).
7 Choi, D. G., Park, G. H. (2018). Analysis of Dispatch Strongpoint for the Fire Accidents Based on Spatial Location-Allocation Model in the Chungnam Province, South Korea, Journal of KARG, 24(2), pp. 267-278.
8 Yun, H. Y., Koo, Y. S., Choi, D. R. (2017). "A Development of Ensemble Model Based on Cluster Analysis to improve PM10 Forecasting Accuracy: Focus on the Weighted Average Ensemble by Weather Cluster.", Journal of Korean Society of Urban Environment, 17(1), pp. 33-42.
9 Song, H. G., Lee, G. H. (2017). Optimal Location Modeling of Early Voting Polls Considering Spatial Accessibility: Cases of Seocho and Gangnam-gu in Seoul, Journal of the Association of Korean Geographers, 52(6), pp. 827-843.
10 Baek, T. H., Son, S. K., Park, B. H. (2016). Modeling Traffic Accident Occurrence Involving Child Pedestrians at School Zone, Journal of Korean Society of Transportation, 34(6), pp. 489-498.   DOI
11 Lee, D. H., Lim, H. S. (2021). Detection of inappropriate advertising content on SNS using k-means clustering technique, Proceedings of the Korea Information Processing Society Conference, pp. 570-573.
12 Kim, J. Y., Lee, H. S., Oh, J. S. (2020). Study on prediction of ship's power using light GBM and XGBoost, Journal of Advanced Marine Engineering and Technology, 44(2), pp. 174-180.   DOI
13 Geron, A. (2017). "Hands-on machine learning with scikitlearn & tensorflow.", 2nd, o'reilly.
14 Jeong, J W., Zhou, K., Oh, S. K., Ryu, B. G. (2021). Design of RBFNN Classifier Using Elastic Net and Its Application, The Korean Institute of Electrical Engineers Conference, pp. 1767-1768.
15 Ke, G., Meng, Q., Finley, T., Wang, T., Chen, W., Ma, W., Ye, Q., Liu, T. (2017). Lightgbm: A highly efficient gradient boosting decision tree, Advances in neural information processing systems, 30, pp. 3146-3154.
16 Lee, H. J. (2020). After installing the smart crosswalk, traffic stop line violations decreased by 70%, https://www.lak.co.kr/m/news/view.php?id=8372 (Dec. 11. 2021).
17 Lee, S., Lee, W. J. (2016). Development of a system for predicting photovoltaic power generation and detecting defects using machine learning, KIPS Transactions on Computer and Communication Systems, 5(10), pp. 353-360.   DOI
18 Choi, H. W., Ko, B. H., Won, D. W., Yeo, H. Y., Yun, W. (2021). The Optimal Location for a Smart Bus Stop through Public Data Analysis: The Case of Changwon-si, Journal of Global Business Research, 33(2), pp. 17-33.   DOI
19 Gan, J. H., Nam, S. H., Seo, Y. J., Jeon, E. J., Lee, K. C. (2019). A Proposal for the Location of Wind Road and Green Space for the Mitigation of Urban Heat Island Effect in Seoul : Focusing on the proposal for heat island index using clustering and variable importance by random forest, Journal of Digital Convergence, 6(1), pp. 27-38.
20 Kim, S. U., Bang, J. I., Hong, S. E., Kim, H. J. (2019). "A Study on the Prediction of Missing Value in Data of Air Pollution Using LightGBM", Korea Institute of Communication Science Conference, pp. 1029-1030.
21 Jang, S. H., Cho, H. E., Jeong, J. W. (2019). Design and Implementation of A Smart Crosswalk System based on Monitoring Vehicle Speed using Deep Learning, Proceedings of the Korean Institute of Information and Commucation Sciences Conference, 23(2), pp. 522-525.
22 Lee, H. J., Lee, S. G. (2021). Comparative Analysis of Machine Learning Models for the Prediction of Pedestrian Crash Severity: Focused on Balancing Pedestrian Crash Dataset, journal of Korean Society for Geospatial Information Science, 29(2), pp. 3-15.
23 Yoon, H. S., Shim, J. Y. (2021). Coffee shop location prediction study using machine learning : Focusing on Coffee Bean case, KMIS International Conference, pp. 183-188.
24 TAAS. (2021), 2019 Traffic Accident Statistical Analysis Report, http://taas.koroad.or.kr/ (Dec. 21. 2021).
25 Jang, S. H., Cho, H. E., Jeong, J. W. (2020). Design and Implementation of A Smart Crosswalk System based on Vehicle Detection and Speed Estimation using Deep Learning on Edge Devices, Journal of the Korea Institute of Information and Communication Engineering, 24(4), pp. 467-473.   DOI
26 Kim, J. I., Chung, H. W., (2001). GIS Applications for Optimum Site Selection of Public Facility, Journal of the KAGIS 4(4), pp. 8-20.
27 Jerome H. Friedman. (2001). Greedy Function Approximation: A Gradient Boosting Machine, The Annals of Statistics, 29(5), pp. 1189-1232.   DOI
28 Kang, W. G., Lee, G. S., Yoon, Y. G. (2016). Factor Analysis of Child-Pedestrian Accidents in Seoul, The KOR-KST Conference, pp. 233-238.
29 Kim, H. B., Kim, S. G., (2006). A Site Selection of Public Facility Based on An Accessibility Theory & GIS Spatial Analysis Technologies, KSCE Journal of Civil Engineering 26(3), pp. 385-391.
30 LightGBM. (2020). Parameters Tuning, https://lightgbm.readthedocs.io/en/latest/Parameters-Tuning.html (Dec. 30. 2021).
31 Lee, G. W., Kim, T. H. (2019). Examining the Characteristics of Traffic Accidents Involving Elderly Drivers in Seoul, South Korea, Ministry of Land, Infrastructure and Transport 102(1), pp. 19-34.
32 Lee, H. S., Kim, H. C., Lee, G. W. (2017). A Study on Priority Determination of the Local Logistics Policies in Chungcheongnam-do Province, Journal of Korea Planning Association, 52(7), pp. 109-120.
33 Lee, S. H., Woo, Y. H. (2018). A Study on the Improvement of Prediction Accuracy for Traffic Accident Models Using Machine Learning (Generalized Regression Neural Network), International Journal of Highway Engineering, 20(6), pp. 179-189.   DOI
34 Ministry of Land, Infastructure and Transport. (2020). standard analysis model, http://www.geobigdata.go.kr/portal/ (Aug. 20. 2021).
35 Na, H. H., Kang, J. H., Kim, D. G. (2021). Deriving the function of a sign for a protected area using the KANO model, Proceedings of the KOR-KST Conference, 85(), pp. 186-186.
36 Pintel. (2021). Pedestrian crossing safety system, http://www.pintel.co.kr/html/index/ (Jan. 11. 2022).
37 Oh, S. T. (2021). Jeollanam-do, Pedestrian-tailored traffic safety policy has an effect, http://m.sisatotalnews.com/article.php?aid=1632627152133515011#_enliple (Nov. 29. 2021).
38 Park, H. G. (2021). "96% of drivers ignored pedestrian crossings without signs", https://m.mk.co.kr/uberin/read.php?sc=30000001&year=2021&no=1086938 (Dec. 29. 2021).
39 Pejman, A., Bidhendi, G. N., Ardestani, M., Saeedi, M., Baghvand, A. (2017). Fractionation of heavy metals in sediments and assessment of their availability risk: A case study in the northwestern of Persian Gulf, Marine pollution bulletin, 114(2), pp. 881-887.   DOI
40 Seoul. (2016). [Smart Crosswalk] Service, https://smart.seoul.go.kr/index.do (Dec. 20. 2021).
41 Seongdong-gu. (2020). Big data analysis for smart crosswalk target selection, https://www.sd.go.kr/main/index.do (Sep. 21. 2021).
42 TAAS. (2018), 2017 Traffic Accident Statistical Analysis Report, http://taas.koroad.or.kr/ (Dec. 20. 2021).
43 TAAS. (2019), 2018 Traffic Accident Statistical Analysis Report, http://taas.koroad.or.kr/ (Dec. 15. 2021).
44 Yeom, M., Kim, K. M. (2011). "Deriving the causes of low fertility and policy demand through cluster analysis.", Journal of Economics Studies, 29(1), pp. 163-190.
45 Yoon, B. J., Ko, E. H., Yang, S. R. (2016). The Study on Traffic Accident of Commercial Vehicle using Odered Logit Model, Proceedings of the Korean Society of Disaster Information Conference, pp. 265-266.