Fuzzy System With The Cooperation Between Visible And Hidden Views

Artificial intelligence technology has developed for nearly 70 years,during which various intelligent methods have been proposed to solve a variety of problems.Fuzzy recognition technology and fuzzy intelligent modeling technology has been extensively concerned and used in healthcare,management,economy and other fields.However,as people's living condition and technology improves,more and more new application scenarios are discovered.Among the new application scenarios,multi-view application scenario has a broad impact on people's life and production.Compared with traditional single view algorithms multi-view algorithms aim at improve the generalization performance by considering multi-view data.Researchers find that the performance of some classical fuzzy system modeling method become no longer reliable in multi-view application scenarios,which facing the following several challenges: In multi-view scenarios,classical fuzzy systems are unable to learn collaboratively from multi-view data since they are proposed in learning scenario of single view.If these classic methods are insistently applied to multi-view learning,they can only be learned in each view independently,which results in unsatisfied outcome.Meanwhile,we found that existing multi-view approaches mainly focus on the views that are visible and ignore the hidden information behind the visible views,which usually contains some intrinsic information of the multi-view data.The relative studies are addressed the challenges that classic fuzzy system modeling methods face in multi-view scenario,and solved the problem that existing multi-view methods unable to make use of the shared hidden information between visible views.First,we proposed a framework of of multi-view fuzzy system with cooperation between visible and hidden views.The proposed framework can make single view fuzzy systems have ability of multi-view learning.Meanwhile,compared with traditional multi-view modeling framework,the proposed framework can make full use of the essential attributes of different views by introducing hidden view information.Secondly,based on non-negative matrix factorization,a method that can extract hidden view information is proposed to exploit the internship between visible views.Finally,using ridge regression extreme learning fuzzy system and classical TSK fuzzy system as base model,we proposed two multi-view fuzzy modeling method with ability of cooperation between visible and hidden views.The experimental results show that the two proposed multi-view methods have better generation ability than single view fuzzy system methods and multi-view algorithms.