Description

Discovering typical medical record extends to give complication for domain applicants. Attributes arranging on any domain derives more time consumption which is only a major problem. In any kind of application data reference for future use or the temporary usage, back end storage is necessary. Multiples of machine learning with data mining techniques ready to solve medical line data stream complexities. Addressing to solve concept-evolution problem and provide a solution that handles all three problems, namely, infinite length, concept-drift, and concept-evolution on new class detection. Aim is to detect novel classes entering automatically even when the classification model is not trained with the novel class instances by using simple classifiers with minimum time taken. Not only detect the new class on using records and also arranging supervised information for eases in the application area. Performance has been improved when novel class overrides the older attributes.