A Review of Fuzzy Cognitive Map Learning Algorithms and Applications
Fuzzy cognitive maps(FCM)are a representative soft computing theory based on cognitive maps and fuzzy set theory.They combine the advantages of both neural networks and fuzzy decision-making and have been successfully applied in many fields,including complex system modeling and time series analysis.Learning the weight matrix is the primary task of modeling based on fuzzy cognitive maps and is the focus of research in this field.To address this core issue,we first comprehensively review the basic theoretical framework of fuzzy cognitive maps and systematically summarize the extended models developed in recent years.Next,the most recent advancements in fuzzy cognitive map learning algorithms are reviewed,analyzed,and summarized.The algorithms are redefined and categorized,with a detailed exploration of their time complexity,strengths,and weaknesses.Additionally,the ap-plication properties of various learning algorithms in various scientific domains are also compared and analyzed in this research,along with the software tools that are now available for creating fuzzy cognitive maps.Finally,poten-tial research directions and development trends for learning algorithms are discussed.
Fuzzy cognitive maps(FCM)learning paradigmcausal inferencesoft computingcomplex system modeling
万方数据
维普
