AN OPTIMIZED MULTI- RELATIONAL CLASSIFICATION MODEL FOR THE PREDICTION OF HEPATITIS

Data mining techniques are commonly applied in different areas such as health-care, banking, manufacturing industry etc.. Medical data are of immense use for predicting patients' health conditions. In health-care, normally, data mining algorithms are used for the analysis of various diseases, to predict the patients at risk of particular diseases and to suggest better medical services at reasonable cost. The present study is designed to provide an efficient multi- relational model for the prediction of ECML/PKDD'05 Hepatitis data. The main idea of this work is to find whether a patient is suffering from hepatitis, if he does what type (B or C) of it and the stages of liver fibrosis (from FO to F4). The classification model was generated based on fuzzy rule-based classifier.