Journal of the Korea Convergence Society (한국융합학회논문지)

Korea Convergence Society (한국융합학회)
권호 표지
DOI QR코드

DOI QR Code

A Methodology for Improving fitness of the Latent Growth Modeling using Association Rule Mining

연관규칙을 이용한 잠재성장모형의 개선방법론

  • Cho, Yeong Bin (Department of Business Administration, Division of International Business, Konkuk Univ.) ;
  • Jun, Jae-Hoon (Department of Biomedical Engineering, College of Biomedical and Health Science, Konkuk University) ;
  • Choi, Byungwoo (Department of Business Administration, Division of International Business, Konkuk Univ.)
  • 조영빈 (건국대학교 국제비즈니스학부 경영학전공) ;
  • 전재훈 (건국대학교 ICT융합공학부 의학공학전공) ;
  • 최병우 (건국대학교 국제비즈니스학부 경영학전공)
  • Received : 2018.11.04
  • Accepted : 2019.02.20
  • Published : 2019.02.28
Download PDF
⟨ Previous Next ⟩

Abstract

The Latent Growth Modeling(LGM) is known as the typical analysis method of longitudinal data and it could be classified into unconditional model and conditional model. It is common to assume that the growth trajectory of unconditional model of LGM is linear. In the case of quasi-linear, the methodology for improving the model fitness using Sequential Pattern of Association Rule Mining is suggested. To do this, we divide longitudinal data into quintiles and extract periodic changes of the longitudinal data in each quintiles and make sequential pattern based on this periodic changes. To evaluate the effectiveness, the LGM module in SPSS AMOS was used and the dataset of the Youth Panel from 2001 to 2006 of Korea Employment Information Service. Our methodology was able to increase the fitness of the model compared to the simple linear growth trajectory.

대표적인 종단자료 분석방법인 잠재성장모형(Latent Growth Modeling)은 무조건적 모형과 조건적 모형으로 구분한다. 잠재성장모형의 무조건적 모형 성장궤적은 선형으로 가정하여 분석하는 경우가 많다. 본 연구는 선형 성장궤적으로 가정하여 모형 적합도가 미달하는 경우 연관규칙기법을 이용하여 모형 적합도를 제고하는 방법론을 제안한다. 방법론은 연관규칙 마이닝의 순차패턴(Sequential Pattern)을 사용한다. 이를 위하여 종단자료를 분위별로 나누고, 각 분위에 속한 종단자료의 기간 변화를 산출한 뒤 이를 순차 패턴 화하였다. SPSS AMOS를 이용하여 한국고용정보원의 2001년부터 6년간 조사한 청년 패널 자료로 효과성을 검증하였다. 기존 단순선형함수를 가정할 때와 비교하여 모형 적합도가 상승하는 것을 확인할 수 있었다.

Keywords

OHHGBW_2019_v10n2_217_f0001.png 이미지

Fig. 1. Longitudinal data characteristics

OHHGBW_2019_v10n2_217_f0002.png 이미지

Fig. 2. longitudinal trajectory of ‘lr_wage’ variable

OHHGBW_2019_v10n2_217_f0004.png 이미지

Fig. 3. Unconditional model of LGM in SPSS AMOS

OHHGBW_2019_v10n2_217_f0005.png 이미지

Fig. 4. 4-fractile 2-sequences association (sequential pattern) rules

OHHGBW_2019_v10n2_217_f0006.png 이미지

Fig. 5. 4-fractile 3-sequences association (sequential pattern) rules

OHHGBW_2019_v10n2_217_f0007.png 이미지

Fig. 6. 4-fractile 4-sequences association (sequential pattern) rules

OHHGBW_2019_v10n2_217_f0008.png 이미지

Fig. 7. 4-fractile 5-sequences and 6-sequences association (sequential pattern) rules

OHHGBW_2019_v10n2_217_f0009.png 이미지

Fig. 8. 5-fractile 2-sequences association (sequential pattern) rules

OHHGBW_2019_v10n2_217_f0010.png 이미지

Fig. 9. 5-fractile 3-sequences association (sequential pattern) rules

OHHGBW_2019_v10n2_217_f0011.png 이미지

Fig. 10. 5-fractile 4-sequences association (sequential pattern) rules

OHHGBW_2019_v10n2_217_f0012.png 이미지

Fig. 12. Conditional Model of LGM

Table 1. Model fitness of simple linear growth trajectory

OHHGBW_2019_v10n2_217_t0001.png 이미지

Table 2. Model fitness of derived sequential patterns vs simple linear pattern

OHHGBW_2019_v10n2_217_t0002.png 이미지

Table 3.Rresult of initial status and slope

OHHGBW_2019_v10n2_217_t0003.png 이미지

Table 4. Model fitness of Conditional Model

OHHGBW_2019_v10n2_217_t0004.png 이미지

References

  1. Hankyoreh (2018.1.22.) http://www.hani.co.kr/arti/international/america/828794.html
  2. E. J. Lee & C. H. Cho. (2013). A Longitudinal Study on the Effects of Franchise's Factors and Performance -Disclosure Agreement, Korean J. of Business Administration, 26(8), 2185-2209.
  3. E. J. Lee & C. H. Cho. (2014). A Longitudinal Study on the Service Quality in Korean Service Industry: Focusing on KS-SQI, Journal of Korea Service Management Society, 15(2), 23-47. DOI : https://doi.org/10.15706/jksms.2014.15.2.002
  4. H. J. Lim & J. S. Cho. (2012). The Effect of Ownership Concentration on Firm Performance : Static and Dynamic Panel Data Analysis, Korean J. of Business Administration, 25(8), 3265-3291.
  5. J. H. Kim. (2018). A longitudinal study of the relationships between commitment type HRM, work team autonomy and innovation performance. The Korean Journal of Human Resource Development Quarterly, 20(2), 1-24. DOI : https://doi.org/10.18211/kjhrdq.2018.20.2.001
  6. Y. B. Cho, S. K. Lee & K. H. Ro. (2015). A Methodology for Analyzing the Longitudinal Data using SOM Technique, Korean J. of Business Administration, 28(1), 93-102.
  7. J. S. Lee & S. Y. Kim. (2017). An Exploration of Nonlinear Latent Growth Model Using Exponential Function: As an Alternative to Quadratic LGM, J. of Educational Evaluation, 30(4), 791-816.
  8. K. S. Kim. (2009). AMOS and LISREL, Han Academy.
  9. Y. B. Cho. (2018). A Data Based Methodology for Estimating the Unconditional Model of the Latent Growth Modeling, J. Digital Convergence, 16(6), 85-93. https://doi.org/10.14400/JDC.2018.16.6.085
  10. S. W. Menard. (2002). Longitudinal research (2nd. ed.). London: Sage Publications Inc.
  11. Toon Taris. (1999). A Primer in Longitudinal Data Analysis, SAGE Publications Inc. DOI : http://dx.doi.org/10.4135/9781849208512.n1
  12. S. S. Yeo & S. H. Park. (2012). An Appliation of Latent Growth Modeling: Use of Curriculum-Based Measurement as longitudinal Data. Asian J. of Education, 13(4), 247-273. DOI :https://doi.org/10.15753/aje.2012.13.4.011
  13. K. L. McArdle & D. B. Epstein. (1987). Latent Growth curves within development structural equation models. Child Development, 58, 110-133. DOI : https://doi.org/10.2307/1130295
  14. B. M. Byrne. (2016). Structural Equation Modeling With AMOS Basic Concepts, Applications, and Programming, Third Edition. New York; Routledge, DOI : https://doi.org/10.4324/9781315757421
  15. R. B. Kline. (2004). Principles and practice of structural equation modeling. New York: Guilford.
  16. K. A. Bollen & P. J. Curran. (2006). Latent curve models: a structural equation perspective. Hoboken, NJ: Wiley-Interscience.
  17. Annie Britton, Yoav Ben-Shlomo, Michaela Benzeval, Diana Kuh & Steven Bell. (2015). Life course trajectories of alcohol consumption in the United Kingdom using longitudinal data from nine cohort studies. BMC Medicine, 13(1), 47. DOI : https://doi.org/10.1186/s12916-015-0273-z
  18. James A. Cranford, Patrick E. Shrout, Masumi Iida, Eshkol Rafaeli, Tiffany Yip & Niall Bolger. (2006). A Procedure for Evaluating Sensitivity to Within-Person Change: Can Mood Measures in Diary Studies Detect Change Reliably? Personality and Social Psychology Bulletin, 32(7), 917-929. DOI :https://doi.org/10.1177/0146167206287721
  19. R Agrawal, T. Imielinski & A. Swami. (1993). Mining association rules between sets of items in large databases, Proceedings of the ACM SIGMOD Conference on Management of Data, 207-216. DOI : https://doi.org/10.1145/170036.170072
  20. B. W. Jin, Y. S. Cho & K. H. Ryu. (2010). Personalized e-Commerce Recommendation System using RFM method and Association Rules. J. of the Korea Society of Computer and Information, 15(12), 227-235. DOI : https://doi.org/10.9708/jksci.2010.15.12.227
  21. J. C. Kim, H. I Jung, H. Yoo & K. Y. Chung. (2018). Sequence Mining based Manufacturing Process using Decision Model in Cognitive Factory. Journal of the Korea Convergence Society, 9(3), 53-59. https://doi.org/10.15207/JKCS.2018.9.3.053
  22. Y. J. Shin & M. S. Yim. (2012). A Study of the Relationship Analysis between Mobile Application by Using An Association Rules. Journal of the Korea Convergence Society, 3(2), 19-25. https://doi.org/10.15207/JKCS.2012.3.2.019
  23. C. G Park & K. E Lee. (2014). A linearity test statistic in a simple linear regression. Journal of the Korean Data and Information Science Society, 25(2), 305-315. DOI : 10.7465/jkdi.2014.25.2.305