Predictive Data Analytics Technique for Optimization of Medical Databases

Abstract

The medical databases are expanding at exponential rate; forfeit technology is required to determine hidden information and facts from such big databases. The data mining technology or Knowledge discovery from databases (KDD) tends to be the well-known technique from past decade which has provided fruitful information to discover hidden patterns and unknown knowledge from large-scale databases. Further, as medical databases are exceeding at an enormous rate data mining tends to be an effective and efficient technology to deal with inconsistent databases which include missing value, noisy attributes, and other types of attributes or factors to discover knowledgeable information for future prognosis of disease. In the current study, we have utilized predictive data analytics technique to diagnose patients suffering from liver disorder. The paper utilizes two-step clustering technology to analyze patients’ disorder with different data variables to find optimal number of clusters of variant shapes and sizes. The focus of study relies on determining hidden knowledge and important factors which can benefit healthcare practitioners and scientists around the globe for prognosis and diagnosis of liver disease at an early stage.