Predictive Big Data Analytics and Privacy based Decision Support System
Lakshita Aggarwal, Puneet Goswami
Abstract
In the era of technological development, Big Data is not simply “business as usual” and the decision to adopt Big Data must take into account many businesses and technology considerations. It’s an accumulation of data that is too large and complex for processing by traditional database management tools. Data retrieved from different sources is analyzed by Operation Analysts or Business Intelligence (BI) analyst’s to take decisions appropriately for desired results. Big data has been pushing current technologies and infrastructure to its limit where the traditional ways of processing really didn’t work anymore. Predictive analysis is one of the buzz areas of research especially in healthcare. Predictive analysis is done to accelerate the diagnostic process before the situation worsens. It will also help in developing new norms and standards to curtail the spread of a pandemic like coronavirus and helps us in caring for those affected by using proper measures against the virus The aim of predictive driver analysis is to provide better facilities and services to the patients and medical practitioners to know what treatment or an asset a person needs based on the study of their past data set. Knowing prior allows our healthcare provider to quickly recommend the right health facility so that correct actions and measures can be taken. This chapter involves a sincere attempt to analyze the predictive analysis of the data sets and then further sort the data sets using some deep learning based tools. Data sets have been analyzed on the basis of V’s of big data. The data of occupancy in different hospitals located at various geo-locations is taken in consideration and is analyzed on the IBM Cognos software to understand the predictive analysis of the model. As the data is accumulated from varied areas therefore its confidentiality was a major concern, so a few techniques have been devised for preserving privacy of the data.