DOI QR코드

DOI QR Code

Future Trend Impact Analysis Based on Adaptive Neuro-Fuzzy Inference System

ANFIS 접근방식에 의한 미래 트랜드 충격 분석

  • 김용길 (조선이공대학교 컴퓨터보안과) ;
  • 문경일 (호남대학교 컴퓨터공학과) ;
  • 최세일 (호남대학교 컴퓨터공학과)
  • Received : 2015.03.12
  • Accepted : 2015.04.23
  • Published : 2015.04.30

Abstract

Trend Impact Analysis(: TIA) is an advanced forecasting tool used in futures studies for identifying, understanding and analyzing the consequences of unprecedented events on future trends. An adaptive neuro-fuzzy inference system is a kind of artificial neural network that integrates both neural networks and fuzzy logic principles, It is considered to be a universal estimator. In this paper, we propose an advanced mechanism to generate more justifiable estimates to the probability of occurrence of an unprecedented event as a function of time with different degrees of severity using Adaptive Neuro-Fuzzy Inference System(: ANFIS). The key idea of the paper is to enhance the generic process of reasoning with fuzzy logic and neural network by adding the additional step of attributes simulation, as unprecedented events do not occur all of a sudden but rather their occurrence is affected by change in the values of a set of attributes. An ANFIS approach is used to identify the occurrence and severity of an event, depending on the values of its trigger attributes. The trigger attributes can be calculated by a stochastic dynamic model; then different scenarios are generated using Monte-Carlo simulation. To compare the proposed method, a simple simulation is provided concerning the impact of river basin drought on the annual flow of water into a lake.

TIA(: Trend Impact Analysis)는 발생될 가능성이 있는 미래의 예기치 못한 사건들을 식별하고 분석하기 위한 고급 예측 도구에 속한다. 적응적인 뉴로-퍼지 추론 시스템은 인공신경망의 일종으로 신경망과 퍼지 로직 원리를 모두 통합하고 보편적 추정되는 것으로 간주한다. 본 논문에서는 적응적인 뉴로-퍼지 추론 시스템을 사용하여 예기치 못한 사건에 관한 심각성의 정도를 추론하고 이를 시간의 함수로서 도입하여 예기치 못한 사건들의 출현 확률에 관해 보다 타당한 추정치를 얻는데 있다. 이러한 접근방식에 대한 배후 개념은 예기치 못한 사건이 갑자기 출현되는 것이 아니라 관련 사건이 가지고 있는 속성 값에 대한 건드림 혹은 변화가 기존 속성 값의 한계를 벗어나 마치 새로운 사건인 것처럼 등장할 수 있음을 전제로 하고 있다. ANFIS 접근 방식은 이러한 사건을 식별해서 예기치 못한 사건의 심각성의 정도를 추론하는데 매우 적절한 방식이라 할 수 있다. 속성들의 변화 값들은 확률적인 동적 모델 및 Monte-Carlo 방법을 사용하여 얻을 수 있다. 제안된 모델에 관한 타당성은 강 유역의 예상치 못한 가뭄에 따른 충격 추세 곡선을 기존 연구 결과와의 비교를 통해 나타낸다.

Keywords

References

  1. T. Gordon, Trend Impact Analysis. "Futures Research Methodology V2," CD ROM, In Proc. the Millennium Project, American Council for the United Nations University, 2003.
  2. N. Agami, A. Omran, M. Saleh, and H. El-Shishiny, "An enhanced approach for trend impact analysis," J. Technological Forecasting and Social Change, vol. 75, no. 9, Nov. 2008, pp. 1439-1450. https://doi.org/10.1016/j.techfore.2008.03.006
  3. D. Suh and C. Park, "A Novel Method of Basic Probability Assignment Calculation with Signal Variation Rate," J. of the Korea Institute of Electronic Communication Sciences, vol. 8, no. 3, 2013, pp. 465-470. https://doi.org/10.13067/JKIECS.2013.8.3.465
  4. L. Firminger, "Trend Analysis : Methods and Problems," Strategic Planning Services. Swinburne University of Technology, TAFE Division, Mar. 2003.
  5. T. Gordon, The Delphi method. "Futures Research Methodology V2," CD ROM, In Proc. the Millennium Project, American Council for the United Nations University, 2003.
  6. N. Agami, M. Saleh, and H. El-Shishiny, "A Fuzzy Logic based Trend Impact Analysis method," J. Technological Forecasting and Social Change, vol. 77, no. 7, Sept. 2010, pp. 1051-1060. https://doi.org/10.1016/j.techfore.2010.04.009
  7. J. Hilbe, Logistic Regression Models. Boca Raton : Chapman & Hall, 2009.
  8. L. Rokach and O. Maimon, Data Mining with Decision Tree : Theory and Applications (Series- Machine Perception and Artificial Intelligence). Singapore : World Scientific Publishing Company, 2008.
  9. J. Berger, Statistical Decision Theory and Bayesian Analysis. New York : Springer-Verlag, 1985.
  10. I. Hwang, "Comparison of confidence intervals for testing probabilities of a system," J. of the Korea Institute of Electronic Communication Sciences, vol. 5, no. 5, 2010. pp. 435-443.
  11. N. Agami, A. Atiya, M. Saleh, and H. El- Shishiny, "A neural network based dynamic forecasting model for Trend Impact Analysis," J. Technological Forecasting and Social Change, vol. 76, no. 7, Sept. 2009, pp. 952-962. https://doi.org/10.1016/j.techfore.2008.12.004
  12. H. Seog and W. Yim, "A Compensation for Distortion of Stero-scopic Camera Image Using Neuro-Fuzzy Inference System," J. of the Korea Institute of Electronic Communication Sciences, vol. 5, no. 3, 2010, pp. 263-268.
  13. E. Lorenz, Nonlinearity, Weather Prediction and Climate Deduction. Oxford : Massachusetts Institute of Technology, Department of Meteorology-Statistical Forecasting Project, 1966.