Preprocessing Methods for Unstructured Healthcare Text Data

Abstract

At present, the amount unstructured text data is increasing exponentially from the past periodically. Information retrieval (IR) from these unstructured text data is challenging. As the data users foresee for particular/specific outcomes. Retrieval of the significant outcomes depends on the fashion, how they are associated/indexed. Unstructured text data like clinical data containing more health information requires challenging preprocessing methods, which also help to reduce the size of the dataset so that it will optimize the performance of the IR system. In this paper, we have proposed the pre-processing methods such as Data collection, Data Cleaning, Tokenization, Stemming, Removal of Stop words which will efficiently help the data users to find the specific patterns from the unstructured text data.