Abstract
In the paper 2D-neo-fuzzy neuron is presented. It is a generalization of the traditional NFN for data in matrix form. 2D-NFN is based on the matrix adaptive bilinear model with an additional fuzzification layer. It reduces the number of adjustable synaptic weights in comparison with traditional systems. For its learning, optimized adaptive procedures with filtering and tracking properties are proposed. 2D-NFN can be effectively used for image processing, data reduction, and restoration of non-stationary signals, presented as 2D-sequences.
Highlights
Artificial neural networks (ANNs) and fuzzy inference systems (FISs) are widely used for solving a large class of data mining tasks, including such modern directions as data stream mining, dynamic data mining, web mining, text mining etc
At the edge of these two approaches, hybrid systems of computational intelligence (HSCI) have emerged, combining the universal approximation properties of ANNs and their ability to learn, and the possibility of linguistic interpretation of the results provided by FIS
HSCI have a number of advantages over ANNs and FISs, their main disadvantage is rather low learning speed provided by gradient algorithms, with the learning rate parameter that is usually chosen from empirical considerations
Summary
Abstract – In the paper, 2D-neo-fuzzy neuron (NFN) is presented. It is a generalization of the traditional NFN for data in matrix form. 2D-NFN is based on the matrix adaptive bilinear model with an additional fuzzification layer. It reduces the number of adjustable synaptic weights in comparison with traditional systems. For its learning, optimized adaptive procedures with filtering and tracking properties are proposed. 2D-NFN can be effectively used for image processing, data reduction, and restoration of non-stationary signals presented as 2D-sequences. Keywords – 2D network, data mining, hybrid systems, neo-fuzzy neuron
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have