Application of the Intuitionistic Fuzzy InterCriteria Analysis Method to a Neural Network Preprocessing Procedure

Abstract

The artificial neural networks (ANN) are a tool that can be used for object recognition and identification. However, there are certain limits when we may use ANN, and the number of the neurons is one of the major parameters during the implementation of the ANN. On the other hand, the bigger number of neurons slows down the learning process. In our paper, we propose a method for removing the number of the neurons without reducing the error between the target value and the real value obtained on the output of the ANN's exit. The method uses the recently proposed approach of InterCriteria Analysis, based on index matrices and intuitionistic fuzzy sets, which aims to detect possible correlations between pairs of criteria.

Two Applications of Inter-Criteria Analysis with Belief Functions

“…This permits to remove all redundant criteria of the original MCDM problem and thus solving a simplified (almost) equivalent MCDM problem faster and at lower computational cost. ICrA has been developed originally by A t a n a s s o v, M a v r o v and A t a n a s s o v a [9], A t a n a s s o v, A t a n a s s o v a and G l u h c h e v [10] and A t a n a s s o v et al [11], based on Intuitionistic Fuzzy Sets [12], and it has been applied in different fields like medicine [13][14][15], optimization [16][17][18][19][20] workforce planning [21], competitiveness analysis [22], radar detection [23], ranking [24][25][26][27], etc. In this paper we improve ICrA approach thanks to belief functions introduced by Shafer in [28] from original Dempster's works [29].…”

InterCriteria Analysis of oncological data of the patients for the city of Burgas

“…The ICA approach has been applied for analyzing data and decision making in different areasmedical investigations [17,18,37,38,40,41], genetic algorithms [1,2,3,21,23,26,29], metaheuristic algorithms [10,11,12,13,14,15,16,19,22,27,28,30,31], neural networks [32,33,34,35], etc.…”

InterCriteria Analysis applied to data from Euro Health Consumer Index for comparing the healthcare systems’ performance in time

“…The results are determined according the following scale ( Figure 1): Figure 1. Scale for determination of the type of the correlations between the criteria ICA is successfully applied in investigations from different science topics: evaluating of university rankings [18,19], neural networks preprocessing procedure [21], genetic algorithms [16,20], the global competitiveness reports [9], chemical research [24,25] and etc. In [26] an application of the ICA approach to data connected with health-related quality of life was presented.…”