A Naïve Bayes Approach for Predicting the Skin Allergy Diseases

Abstract

AbstractMachine learning (ML) plays a significant part in data mining, image identification, natural language processing, and disease diagnosis in medical industry. ML provides precise outputs in many fields discussed above (Geetha and Arunachalam in Evaluation based approaches for liver disease prediction using machine learning algorithms. Dr. MGR University, Chennai, India, 2021 [1]). This research study has analyzed various diseases such as impetigo, fungal infections, allergy, and other diseases. For analyzing the data, ML techniques are used to enhance over time and work efficiently but models require error-free data (Saboji and Ramesh in A scalable solution for heart disease using classification mining technique. CMR Institute of Technology, India, 2017 [2]). The researches have previously utilized different models to solve different challenges in the field of healthcare and diagnosis. The maximum accuracy is shown by naïve Bayes (Snehith Raja and Anurag in Machine learning based heart disease prediction system. Vardhaman College of Engineering, India, 2021 [3]). The algorithm that can compete with various precise models is random forest as it delivers a highly error-free prediction. Therefore, the proposed application requires highly precise results and it does not need a human readable model, which can make use of random forest (RF) algorithm (Chowdhury and Ahmed in Heart disease prognosis using machine learning classification techniques. Metropolitan University, Bangladesh, 2021 [4]). The decision tree algorithm delivers the second highest accuracy rate. This algorithm is convenient and operate fast forecastable class of test data.KeywordsMachine learning (ML)K-nearest neighbor (KNN)K-meansLogistic regression (LR)Decision tress classifier (DSC)Random forest classifier (RFC)Naïve Bayes (NB)Convolutional neural network (CNN)Recurrent neural networks (RNN)Support vector machine (SVM)Artificial neural network (ANN)