An Efficient Algorithm for Disease Diagnosis Using Hybrid Fuzzy-rough Set Model

The model presented herein is an information system which guarantees avoidance of redundancy besides minimizing complexities in computing data. This could be used in creating rules which may serve as an aid in elucidation of our knowledge in fields such as medicine. In situations when the information system possesses redundancy, it is necessary to treat data in any one of the ways using the concept of reduction without altering the indiscernible relations. Data reduction concept aims to determine a minimal data subset from a problem domain while retaining suitable original data. In this context fuzzy set and rough set (RST) theories are being used as a mathematical tool to perform data reduction and framing decision rules as a pre-processing with little success. In the existing system, some of the attributes may be lost and not providing required accuracy of attributes. Hence, in the present work, by employing a hybrid variant of fuzzy set and rough set, a new reduction technique called ad Fuzzy-Rough set model is proposed. This would be use as an efficient algorithm for the reduction of information set without any loss and framing the decision rule to be employed successfully efficiently for disease diagnosis in medical field.