An Efficient Healthcare Medication System with Clustering Algorithm Using Euclidean Distance Adjoining Data Lake

The health care services make use of the patient information along with big data analytics towards the doctor's prescription advice to patients. Still, the majority medical data is not in the standard structured standard format. So it requires enormous effort in converting the unstructured data to structured format. This paper highlights a data lake architecture which helps in decreasing the data ingestion time thereby increasing and improving the accuracy in healthcare data analytics. This facilitates the analytical process to be done completed with the removal of data silos. The new intended data lake architecture improves working analysis of data in the way of making connections with third party data providers. The Hadoop Distributed File System (HDFS) is used for storing the structures as well as unstructured data. This paper utilizes the Euclidean distance for clustering process where the patients with similar related health conditions are clustered. Then each cluster is subjected to Support Vector Machine (SVM) for the accurate doctor's prescription. The experimental results showed a decrease in ingesting data by the proposed data lake architecture. One more advantage is the precise and accurate cluster formation than the existing methods. It also improves the healthcare medication by eliminating the data silos.