Forecasting Emergency Department in Hospitals Using Regression Prediction Models

In this work, an efficient and scalable forecasting methodology developed for accurately predicting the emergency department according to the volume of patients in hospitals. It is accomplished with the help of Google Trends data, which is publicly available that helps to retrieve the volume of data with respect to the keyword queries. Here, the correlation and regression analysis are performed for improving the performance of prediction, where supports to simplify the process of data retrieval. Typically, the Google Trends is a kind of searching tool and has been extensively used for tracking the searching items with different queries of information. Also, it is specifically used for improving the predicting analytics that highly supports the operations of blogging, mapping, and streaming. In the recent days, overcrowding is one of the major problems in the emergency departments of hospitals, hence it is considered as the ultimate threat of emergency medical care systems. It is mainly associated with increased errors of healthcare professionals, delay time care, and more deaths rate. The main contribution of this work is to utilize the internet searching data that is publicly available in the Google Trends for exactly predicting the patient volume of emergency department. Here, the correlation analysis methods have been used for identifying the relationship among the patient volume and predictors obtained from the search data.