A Comprehensive Review on Machine Learning Models for Medical Data Classification

Abstract

The content of multi-modal medical features representing medical data was researched and many modes of learning were analyzed for the risk assessment of diseases in order to gather information from medical health data and generate intelligent application relevant inquiries. The representation of medical data employs the method of deep learning. To attain model adaptability, the study and extraction of different disease features employ the same procedure. Classification of medical data is an important data mining challenge that has interested various academics throughout the world for a decade. Various data mining algorithms are utilized for health data sorting such as classification, clustering, regression, soft computing and more. Feature selection algorithms play a key part in every challenge in machine learning. The selection of the best method produces an optimal subset of properties, which enhances accuracy and reduces training time. It is also beneficial to remove the useless characteristics for high-dimensional datasets. This paper provides a brief review on medical data classification models to provide a brief assessment of the common feature selection and classification models specifically utilized for medical data classification. Various aspects, such as the medical data sets available, selection of features, choice of classification, problems in the identification and analysis of the key methods of selecting features and detailed mechanisms are represented in this survey paper.