A modified ANFIS model for classification of blood donors

Abstract

The aim of this research paper is to develop a modified adaptive neuro fuzzy inference system based gradient descent algorithm (MANFIS-GD) model for blood donor classification. This paper further describes a methodology for developing the proposed model by incorporating the capability of fuzzy logic (FL) and artificial neural network learning algorithm, gradient descent (GD) algorithm. Indexing unique membership functions in a row-wise vector using a novelty vectorisation technique in adaptive neuro fuzzy inference system (ANFIS) algorithm to obtain efficient classification performance results and make faster convergence rate was introduced. The proposed model maximises the correctly classified data and minimise
the number of incorrectly classified patterns. In other words, the MANFIS-GD model was able to correctly predict classification labels of donors who have denoted blood and those who do not denote blood. In addition, an attempt was done to specify the effectiveness of the performance measuring classification accuracy, sensitivity and specificity. In comparison, the proposed method achieves superior performance when compared to conventional method(ANFIS-GD) and to some related existing methods.