Nodes Selection Criteria for Fuzzy Cognitive Maps Designed to Model Time Series

The article introduces three concepts’ rejection/selection criteria for Fuzzy Cognitive Map-based method of time series modeling and prediction. Proposed criteria are named entropy index, membership index and global distance index. Concepts’ selection strategies facilitate Fuzzy Cognitive Map design procedure. Proposed criteria allow to simplify, otherwise very complex models, and achieve a reasonable balance between complexity and accuracy.

[1]  Chunyan Miao,et al.  An Extension to Fuzzy Cognitive Maps for Classification and Prediction , 2011, IEEE Transactions on Fuzzy Systems.

[2]  Xiaodong Liu,et al.  The linguistic forecasting of time series based on fuzzy cognitive maps , 2013, 2013 Joint IFSA World Congress and NAFIPS Annual Meeting (IFSA/NAFIPS).

[3]  Elpiniki I. Papageorgiou,et al.  Extended Evolutionary Learning of Fuzzy Cognitive Maps for the Prediction of Multivariate Time-Series , 2014, Fuzzy Cognitive Maps for Applied Sciences and Engineering.

[4]  Elpiniki I. Papageorgiou,et al.  Application of evolutionary fuzzy cognitive maps to the long-term prediction of prostate cancer , 2012, Appl. Soft Comput..

[5]  Bart Kosko,et al.  Fuzzy Cognitive Maps , 1986, Int. J. Man Mach. Stud..

[6]  Witold Pedrycz,et al.  Numerical and Linguistic Prediction of Time Series With the Use of Fuzzy Cognitive Maps , 2008, IEEE Transactions on Fuzzy Systems.