The Journal of Korean Institute of Information Technology (한국정보기술학회논문지)

Korean Institute of Information Technology (한국정보기술학회)
권호 표지
DOI QR코드

DOI QR Code

Demand Forecasting for G2B E-commerce Using Public Data : A Case Study of Public Procurement Service

공공데이터를 이용한 G2B 전자상거래 시장수요예측 - 조달청 사례를 중심으로 -

  • 박현기 (서울과학기술대학교 IT정책전문대학원 산업정보시스템공학과) ;
  • 안재경 (서울과학기술대학교 글로벌융합산업공학과)
  • Received : 2014.08.14
  • Accepted : 2014.10.01
  • Published : 2014.10.31
⟨ Previous Next ⟩

Abstract

In this paper, an ARIMA based demand forecasting program has been implemented by utilizing public data on G2B e-commerce at public procurement service. Recently, there has been growing interest in research and use of public data. However, it is found that few studies proposed systematic procedures for collecting, processing, and analysing the public data. There also have been limitations on grasping the attributes of the public data, extracting them for researchers' purpose and applying the right methodologies. This study presents a series of procedures for redefining and extracting the public data, and attempts to forecast the market sizes in each segment using the ARIMA model. This paper also sheds light on utilizing the public data in the area of demand forecasting by programming the whole processes where unit root test, model identification, model estimation, and model diagnostic checking are automatically to be done. As a result of the analysis, MAPE values fall into 3.90%~24.47%, which shows that the ARIMA model implemented in this paper is working well.

본 논문은 조달청 종합쇼핑몰 공공데이터를 체계적으로 활용하여 ARIMA 모형 기반의 시장수요예측 프로그램을 구현하였다. 최근 공공데이터의 활용과 연구에 대한 관심이 높아지고 있으나, 공공데이터의 수집, 가공 및 분석을 체계적으로 수행한 연구는 제한적인 실정이다. 공공데이터를 활용하여 분석하기 위해서는 관련데이터의 속성을 구체적으로 파악하여 사용자의 의도에 맞도록 데이터를 추출하여야 하며, 이를 토대로 연구하고자 하는 방법론을 적용하여야 하지만, 현재까지의 연구에서는 이를 처리하는 데 한계가 있었다. 본 연구는 이러한 한계점을 보완하고자 공공에서 제공하는 데이터의 속성을 파악하고 추출하였으며, 프로그래밍을 통하여 일괄적으로 부문별 시장규모를 예측할 수 있는 방법을 제시하였다. ARIMA 모형의 단위근 검정, 모형식별, 모형추정 및 모형검증에 대한 전 과정을 프로그래밍화하여 일괄적으로 처리하고 예측함으로써 공공데이터의 처리와 분석이 가능함을 제시하였다. 분석결과 MAPE 값이 3.90%~24.47%로 본 연구에서 구현한 ARIMA 모형을 이용한 처리방법이 우수한 것으로 나타났다.

Keywords

Acknowledgement

Supported by : 서울과학기술대학교

References

  1. Public data portal, http://www.data.go.kr
  2. Appleford, Simon, James R. Bottum, and Jason Bennett Thatcher, "Understanding the social web: towards defining an interdisciplinary research agenda for information systems", The DATABASE for Advances in Information Systems, Vol. 45, No. 1, pp. 29-37, Feb. 2014. https://doi.org/10.1145/2591056.2591059
  3. Seok-Bong Jeong, "A Novel Approach to Customer Classification and Preference Goods Extraction based on Social Network Analysis for Internet Shopping Malls", Entrue Journal of Information Technology, Vol. 13, No. 1, pp. 57-68, Apr. 2014.
  4. Dae-Gi Kim, Won-Kyun Joo, Eun-jin Kim, and Yong-Ho Lee, "A Case Study on Classification System Design for Public Sector Information Typology", Journal of Digital Policy & Management, Vol. 12, No. 4, pp. 51-68, Apr. 2014.
  5. Box, George EP, and Gwilym M. Jenkins. Time series analysis: forecasting and control, revised ed. Holden-Day, 1976.
  6. Sang-Yun Lee and Hong-Joo Yoon, "The Study on Strategy of National Information for Electronic Government of S. Korea with Public Data analysed by the Application of Scenario Planning", The Journal of the Korea institute of electronic communication sciences, Vol. 7, No. 6, pp. 1245-1258, Dec. 2012.
  7. Yong-Jun Cho, Yong-Hoon Cho, and Jeong-Uk Kim, "Time Series Analysis of General Marine Fisheries,Journal of Rural Development", Vol. 29, No. 1, pp. 123-134, June 2006.
  8. Keun-Soon Kim and Jae-Hyun Ahn, "A Study on the Real Time Forecasting for Monthly Inflow of Daecheong Dam using Seasonal ARIMA Model", Journal of Korea Water Resources Association, pp. 1395-1399, May 2010.
  9. Pasapitch Chujai, Nittaya Kerdprasop and Kittisak Kerdprasop, "Time Series Analysis of Household Electric Consumption with ARIMA and ARMA Models", Proc. IMECS Conf., Hong Kong. Mar. 2013.
  10. Young-Joo Kim, "Study on Low Cost Carrier Demand Forecasting Using Seasonal ARIMA Model", Journal of Tourism Studies, Vol. 26, No. 1, pp. 3-25, Feb. 2014.
  11. Hee-Jae Kim, "The Application of Time Series Analysis under R Environment", Journal of the Korean Data Analysis Society, Vol. 13, No. 1, pp. 331-341, Jan. 2011.
  12. Kyo-Won Choi, "A Study on the Prediction Methods of Domestic e-Commerce Market Size", The Journal of Society for e-Business Studies, Vol. 9, No. 4, pp. 1-17, Nov. 2004.
  13. Man-jai Lee, "Big Data and the Utilization of Public Data", Internet and Information Security, Vol. 2, No. 2, pp. 47-64, Nov. 2011.
  14. The Korea Chamber of Commerce and Industry, http://www.gs1kr.org
  15. Ig-Hoon Lee, "A Study on Supplier Relationship Management System for National Public Procurement", The Journal of Society for e-Business Studies, Vol. 16, No. 1, pp. 101-116, Feb. 2011. https://doi.org/10.7838/jsebs.2011.16.1.101
  16. Dong-soo Lee, "Forecasting Quarterly Growth Rates Using IAIP", Journal of The Korean Official Statistics, Vol. 18, No. 2, pp. 66-88, Oct. 2013.
  17. Hyun-ki Park and Jae-kyoung Ahn, "Implementation of the Demand Forecasting Module Linking JAVA and R Programs", Journal of Korean Institute of Information Technology, Vol. 12, No. 3, pp. 85-94, Mar. 2014.
  18. Sang-Kuk Kim, "The Role of Agricultural Cooperative under the Forecast of Korean Apiculture by the ARIMA Model", The Korean Journal of Cooperative Studies, Vol. 25, No. 1, pp. 183-210, Sep. 2007.
  19. Sang-Sok Suh, Jong-Woo Park, Gwang-suk Song, and Seung-Gyun Cho,"A Study of Air Freight Forecasting Using the ARIMA Model", Journal of Distribution Science, Vol. 12, No. 2, pp. 59-71, Feb. 2014. https://doi.org/10.15722/jds.12.2.201402.59