A Novel Data Mining Based Approach for Healthcare Applications

Abstract

Data mining has been bolstered by a massive increase in the usage of data analysis in all sectors. The greatest significant influence on a person's quality of life today is access to quality healthcare. A patient's health might be adversely affected by sudden changes in short-term patterns. It is extremely difficult to interpret the large amounts of data created by health institutions to make significant decisions about patient health. Personal healthcare is out of reach due to the stress of the job. Clustering techniques like K-means and D-stream are used to determine whether a person is fit or unfit based on their historical and real-time data. Both clustering techniques are used on the biological history database of the patient in question. We tested both algorithms using real-world biological data to ensure they were accurate. The D-stream method, a density-based clustering technique, addresses the K-means algorithm's shortcomings. Finally, we can determine the efficacy and efficiency of both algorithms by computing their performance metrics.