Year-end Sale % OFF 
Abstract
Representing and analyzing the complexity of models constructed by data is a difficult and challenging task, hence the need for new, more effective techniques emerges, despite the numerous methodologies recently proposed in this field. In the present paper, the main idea is to systematically create a nested structure, based on a fuzzy cognitive map (FCM), in which each element/concept at a higher map level is decomposed into another FCM that provides a more detailed and precise representation of complex time series data. This nested structure is then optimized by applying evolutionary learning algorithms. Through the application of a dynamic optimization process, the whole nested structure based on FCMs is restructured in order to derive important relationships between map concepts at every nesting level as well as to determine the weights of these relationships on the basis of the available time series. This process allows discovering and describing hidden relationships among important map concepts. The paper proposes the application of the suggested nested approach for time series forecasting as well as for decision-making tasks regarding appliances’ energy consumption prediction.
Highlights
In recent years, fuzzy cognitive maps (FCMs) have become increasingly popular [1]
In Reference [10], a novel approach based on FCMs and a granular fuzzy set-based model of inputs were proposed for realizing time series prediction at the linguistic and numerical levels
The simple average method was used to calculate the forecasted values of the output concept based on values generated by the FCM models belonging to the second level
Summary
Fuzzy cognitive maps (FCMs) have become increasingly popular [1]. An FCM can be defined as a type of recurrent neural network, carrying the main aspects of fuzzy logic. The paper recommends applying the proposed nested approach for time series forecasting as well as for decision-making tasks in the field of appliances’ energy consumption prediction. In Reference [14], historical time series involving energy consumption data were utilized along with the application of the Structure Optimization Genetic Algorithm (SOGA) [12] for the automatic construction of an evolutionary FCM. A two-stage prediction model for forecasting was introduced in Reference [14], which exploits the competent characteristics of evolutionary FCMs enriched with those of SOGA and ANNs. Recently, in Reference [13], a new ensemble-based forecasting approach was proposed for time series analysis, which deals with the problem of natural gas demand prediction, case studying three major cities in Greece.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
