Blood Bank Clustering: Improving Performance of Clustering using Entropy Weighted K-Means

This research provides a novel method for improving clustering efficiency in blood banks. Blood transfusions are used to save the lives of billions of people throughout the world. Every day, about 2 million people are involved in accidents throughout the world, resulting in an emergency scenario in which we require adequate blood for the patient to cover up the blood loss and ensure his survival. Blood banks playa significant role in this situation; a blood bank is a spot where blood is collected and stored for future use; nevertheless, locating a suitable blood bank in the surrounding area is a common task for patient caretakers. In this research, we presented an effective clustering or grouping of blood banks to tackle this problem. Our primary objective here is to first cluster blood banks using an advanced version of the k means clustering algorithm, namely the Entropy weighted k means clustering algorithm while taking into account factors such as longitudinal and latitudinal points of blood banks, blood bank category (Government, Charity, and Private), and blood component availability. We use the sum of the square error to assess the effectiveness of this clustering algorithm. In this process, we also identify the region where a new blood bank is needed based on longitudinal and latitudinal information.