Automated Classification of Health Records for Disease Prediction Using NLP and Machine Learning

Abstract

Healthcare industry efficiency and accuracy can be enhanced through disease prediction when using such electronic health records to extract meaningful knowledge. Applying structured and unstructured medical information enables the system to generate improved health predictions from EHR data. The clinical text preparation uses natural language processing alongside three main machine learning models including Support Vector Machines (SVM), Random Forest and Neural Networks along with other algorithms for classification functions. The system utilizes public health data to measure its accuracy performance through precision, recall and F1 scores evaluation.