DOI QR코드

DOI QR Code

최적화된 Gradient-Boost를 사용한 서울 자전거 데이터의 결정 요인 예측

Predicting Determinants of Seoul-Bike Data Using Optimized Gradient-Boost

  • 김차영 (경기대학교 교양학부) ;
  • 김윤 (국립한국복지대학교 컴퓨터정보보안과)
  • 투고 : 2022.10.05
  • 심사 : 2022.11.01
  • 발행 : 2022.11.30

초록

서울시에서는 공유 자전거 시스템, "따릉이"를 2015년부터 도입, 운영하여, 교통량 감축과 대기오염 해소를 위해 노력하고 있다. 하지만 공유 자전거 시스템, "따릉이"의 운영전략 미훕으로 인해 많은 문제가 발생하고 있어 이를 해결하고자 다양한 연구들이 제시되고 있다. 이들 연구의 대다수는 수요와 공급의 불균형을 해결하고자 하는 전략적 "자전거 배치"에 집중되어 있으며 또한 이들 중 다수가 날씨나 계절과 같은 특징을 그룹화함으로써 수요를 예측하고 있다. 그리고 이전에는 이들 예측방법이 주로 시계열 분석을 기반으로 하고 있었으나 최근에는 딥러닝/머신러닝으로 수요를 예측하는 연구들이 속속 등장하고 있다. 본 논문에서는 기존에 제시된 다양한 특징들을 기반으로 하면서, 새로운 특징을 발견하고 선택된 특징들의 중요도를 비교, 이를 순서화함으로써, 보다 정확한 수요 예측이 가능함을 보인다. 그리하여, 우리는 기존의 딥러닝/머신러닝 및 시계열 분석을 그대로 사용하면서 비교적 정확한 결정계수를 획득하고 이를 이용해 개선된 수요예측이 가능하도록 한다.

Seoul introduced the shared bicycle system, "Seoul Public Bike" in 2015 to help reduce traffic volume and air pollution. Hence, to solve various problems according to the supply and demand of the shared bicycle system, "Seoul Public Bike," several studies are being conducted. Most of the research is a strategic "Bicycle Rearrangement" in regard to the imbalance between supply and demand. Moreover, most of these studies predict demand by grouping features such as weather or season. In previous studies, demand was predicted by time-series-analysis. However, recently, studies that predict demand using deep learning or machine learning are emerging. In this paper, we can show that demand prediction can be made a little better by discovering new features or ordering the importance of various features based on well-known feature-patterns. In this study, by ordering the selection of new features or the importance of the features, a better coefficient of determination can be obtained even if the well-known deep learning or machine learning or time-series-analysis is exploited as it is. Therefore, we could be a better one for demand prediction.

키워드

참고문헌

  1. E. LEE and B. SON, "Optimal Rebalancing Strategy for Public Bike-sharing System in Seoul", J. Korean Soc. Transp. Vol.37, No.1, pp. 28-38, 2019. https://doi.org/10.7470/jkst.2019.37.1.027
  2. C. Chatfield, "The Holt-Winters Forecasting Procedure," Journal of the Royal Statistical Society, Series C(Applied Statistics), Vol.27, No.3, pp.264-279, 1978. https://doi.org/10.2307/2347162
  3. K. CHO. S. S. Lee, and D. H. Nam, "Forecasting of Rental Demand for Public Bicycles Using a Deep Learning Model," J. Korea Inst. Intell. Transp. Syst., vol.19, no.3, pp.28-37, 2020. https://doi.org/10.12815/kits.2020.19.3.28
  4. S. Park, M. Kim, and J. Im, "Estimation of Ground-level PM10 and PM2.5 Concentrations Using Boosting-based Machine Learning fromSatellite and Numerical Weather Prediction Data," Korean Journal of Remote Sensing, Vol.37, No.2, pp.321~335, 2021. https://doi.org/10.7780/kjrs.2021.37.2.11
  5. T. Park and C. Kim, "Predicting the Variables That Determine University (Re-)Entrance as a Career Development Using Support Vector Machines with Recursive Feature Elimination: The Case of South Korea," Sustainability, MDPI, Vol.12, No.18, pp.1-11, 2020. https://doi.org/10.3390/su12187365
  6. H. Lim and K. Chung, "Development of Demand Forecasting Model for Seoul Shared Bicycle," Jour. of KoCon.a, vol.19, no.1, pp.132-140, 2019. https://doi.org/10.5392/JKCA.2019.19.01.132
  7. Kyung-Ok Kim, "A Study on the Characteristics of Shared Bike Use for Operation of a Shared Bike System Considering Bicycle Imbalance," 2018 Seoul Research Paper Competition, Seoul Research Institute, 16th 2018. https://www.si.re.kr/node/61093
  8. S. M. Woo., G. Y. Kim. and H. C. Kim, "How to Improve Suitability of Irradiation Utilization in Development of Linear Regression Model for Estimating Paprika Productivity", The Journal of the Convergence on Culture Technology (JCCT), Vol. 7, No. 4, pp.779-783, November 30, 2021. https://doi.org/10.17703/JCCT.2021.7.4.779
  9. C. Kim, and T. Park, "Predicting Determinants of Lifelong Learning Intention Using Gradient Boosting Machine (GBM) with Grid Search," Sustainability MDPI, Vol.14, No.9, Article Number. 5256, 2022. https://doi.org/10.3390/su14095256
  10. S. H. Moon, "Analysis of AI-Applied Industry and Development Direction," The Journal of the Convergence on Culture Technology (JCCT), Vol. 5, No. 1, pp.77-82, February 28, 2019. https://doi.org/10.17703/JCCT.2019.5.1.77
  11. J Son, C. Kim, and M. Jeong, "Unsupervised Learning for Anomaly Detection of Electric Motors," Vol.23 Issue 4, pp.421-427, April 2022. http://doi.org/10.1007/s12541-022-00635-0