Complex fuzzy computing to time series prediction: a multi-swarm PSO learning approach

Authors:
Chunshien Li;Tai-Wei Chiang
Affiliations:
Department of Information Management, National Central University, Taiwan, ROC;Department of Information Management, National Central University, Taiwan, ROC
Venue:
ACIIDS'11 Proceedings of the Third international conference on Intelligent information and database systems - Volume Part II
Year:
2011

Citing 10
Cited 4

Multilayer feedforward networks are universal approximators

Neural Networks
Fuzzy complex numbers

Fuzzy Sets and Systems
Inferring operating rules for reservoir operations using fuzzy regression and ANFIS

Fuzzy Sets and Systems
Multigrid-based fuzzy systems for time series prediction: CATS competition

Neurocomputing
Incremental learning of dynamic fuzzy neural networks for accurate system modeling

Fuzzy Sets and Systems
Fuzzy prediction architecture using recurrent neural networks

Neurocomputing
An Adaptive Network-Based Fuzzy Inference System (ANFIS) for the prediction of stock market return: The case of the Istanbul Stock Exchange

Expert Systems with Applications: An International Journal
Complex fuzzy sets

IEEE Transactions on Fuzzy Systems
Complex fuzzy logic

IEEE Transactions on Fuzzy Systems
Toward complex fuzzy logic

IEEE Transactions on Fuzzy Systems

A novel self-organizing complex neuro-fuzzy approach to the problem of time series forecasting

Neurocomputing
Efficient differential evolution algorithm-based optimisation of fuzzy prediction model for time series forecasting

International Journal of Intelligent Information and Database Systems
Efficient differential evolution algorithm-based optimisation of fuzzy prediction model for time series forecasting

International Journal of Intelligent Information and Database Systems
A granular neural network: Performance analysis and application to re-granulation

International Journal of Approximate Reasoning

Quantified Score

Hi-index	0.00

Visualization

Abstract

A new complex fuzzy computing paradigm using complex fuzzy sets (CFSs) to the problem of time series forecasting is proposed in this study. Distinctive from traditional type-1 fuzzy set, the membership for elements belong to a CFS is characterized in the unit disc of the complex plane. Based on the property of complex-valued membership, CFSs can be used to design a neural fuzzy system so that the system can have excellent adaptive ability. The proposed system is called the complex neuro-fuzzy system (CNFS). To update the free parameters of the CNFS, we devise a novel hybrid HMSPSO-RLSE learning method. The HMSPSO is a multi-swarm-based optimization method, first devised by us, and it is used to adjust the premise parameters of the CNFS. The RLSE is used to update the consequent parameters. Two examples for time series foresting are used to test the proposed approach. Through the experimental results, excellent performance has been exposed.