Machine Learning Algorithms for Medicine Text Mining Disease Diagnosis

The proliferation of digital devices like smartphones and the Internet of Things has resulted in a flood of data. Every second, millions of bytes of data (including text, music, images, and videos) are produced. Thus, the need for big data analytics has been sparked by the current situation. Healthcare service records are an example of the large amounts of data created every day by hospitals. Information on patient symptoms, drug use, and the discovery of new illnesses, among other things, may be gleaned from this massive dataset. In this investigation, we use text processing methodology to public hospital patient record text data from 2017–2019 to better understand how diseases manifest in their patients. Preprocessing data from medical records involves using the Nave Bayes Classifier & Random Forest algorithms to categorise illnesses. To narrow down the collection of symptom statements, a list of altered Indonesian stop words was employed. Compared to the Naive Bayes classification technique. Experiments validate the effectiveness and efficiency of our suggested technique for illness classification in medical records.