Critical Evaluation of Predictive Analytics Techniques for the Design of Knowledge Base

Abstract

The present diagnosis methods in medical fields are aided very much by the cluster analysis methods. Data Summarization techniques are used to discover the hidden patterns in huge datasets. They may be used for future interpretation in diverse aspects in different environments. In the context of medical data bases, the enormous growth of medical information and its corresponding use for disease diagnosis is a strenuous process. Therefore Disease diagnose systems requires the conventional data analysis combined which proficient knowledge of different diseases. Recent developments in Data segmentation techniques may be used to analyze the reports of the liver patients together with trends of the diseases and standard processes for resource utilization in health care problems. Development of new system based on the above analysis in turn assist the physician for better diagnosis of disease. In the present paper, various classification techniques are applied to predict the disorders in the liver functions accurately. The present paper is aimed at proposing a new method for the prediction of the diseases with a better accuracy than the existing traditional classification algorithms. It was found that these results are very much promising and more accurate.