Hybrid Multi-Cloud based Disease Prediction Model for Type II Diabetes

Advancements in health informatics pave the way to explore new medical decision making systems which are characterized by an exponential evolution of knowledge. In the medical domain, disease prediction has become the centre of research with the increasing trend of healthcare applications. The predictive knowledge for the diagnosis of disease highly depends on the subjective knowledge of the experts. So the development of a disease prediction model in time is essential for patients and physicians to overcome the problem of medical distress. This paper explores a hybrid approach (Cooperative Ant Miner Genetic Algorithm) for classifying the medical data. Three benchmarked Type II diabetic datasets (US, PIMA, German) from the UCI machine learning repository were used to analyze the effectiveness of the disease prediction model. The devised classification algorithm with a Soft-Set approach was deployed in a Multi-Cloud environment for enhancing the storage and retrieval of data with reduced response and computation time. The cooperative classification algorithm in the cloud database distinguishes the diseased cases from the normal ones .The soft set theory analyzes the severity of the diseased cases by calculating the percentage of diabetic risk using soft intelligent rules and stores them in a separate knowledge base. Thus the proposed model serves as a suitable tool for eliciting and representing the expert’s decision which aids in prediction of Type II diabetic risk percentage leading to the timely treatment of patients.