Stacking Ensemble Model for Pediatric Gastroenteritis Disease Classification Using Structured and Unstructured Dataset

Abstract

With the progressive advancement in the field of IT approximately 2.5 newlinequintillion bytes of data are generated every day. The data are available in the form of newlinestructured, unstructured and semi-structured data. Out of all three types of data, unstructured newlinedata will account for 90% of the total data produced. In Medical domain, unstructured data newlineis produced through various resources such as clinical notes, discharge summaries, lab newlinereports and so on, in large amount. Such unstructured data has to be extracted and analyzed newlineto produce useful information like disease classification, disease stage identification, etc. newlineThe proposed research work aims at developing a system which can address the challenges newlinesuch as feature extraction from unstructured data, filling the missing values in the extracted newlinedata records and handling data imbalance. These challenges are addressed so that the disease newlineclassification model will be trained better to accurately classify the disease and identify the newlinestage of the disease newline