Predicting Modalities of Dyslexic Students using Neuro-Linguistic Programming to Enhance Learning Method

Dyslexia causes difficulty in reading, writing and learning. The children at a tender age have always suffered due to dyslexia. Dyslexia deceives student's perception and makes it difficult in the process of learning. In this paper, machine learning techniques like multi-layer perceptron, Decision tree and Gaussian NB approaches were implemented for the prediction of modalities. To enhance the learning approach for the students suffering from dyslexia, the predicted modalities can be adopted. The sampled data was trained, and the target labels were classified into three classes as visual, auditory, and kinesthetic. The data was processed and fed into multi-layer perceptron, decision tree and naive bayes machine learning algorithms using scikit-learn. Confusion matrix was used to evaluate the performance measure of the algorithms. It was observed that models achieved accuracy of 81.41% for MLP Classifier, 63.82% for Decision tree and 79.25% for Naive bayes. The best result was achieved by MLP Classifier.