Extracting Reliable Health Condition and Symptom Information to Support Machine Learning

Machine Learning (ML) technologies in recent times are widely applied in various areas to assist knowledge gaining and decision-making tasks and healthcare is one of the important area among these tasks. In this paper, we propose a process to identify reliable health data from online resources and process the data to enable being used by the ML technologies. As an example, we scrap a condition-symptom dataset with Natural Language Processing (NLP) features from one of the UK NHS website. In addition, we examine our data in depth by having symptom frequency, similarity and clustering analysis.