Effective classification of diabetes using big data analytics

Diabetes Miletus (DM) is a non-communicable disease which has affected more people in India. According to the recent survey, Diabetes Miletus stands at fourth place in the world with India alone accounting to around 50 million. Diabetes Miletus is classified as Type 1 and Type 2 diabetes respectively. This disease may prolong for decades and consequently lead to chronic complications such as foot ulceration, neuropathy, retinopathy and nephropathy. Hospitals produce huge amount of patient data which is stored in the database in a structured or unstructured form. This data must be analyzed using automated tools to extract the knowledge which can be used to classify the diabetic data of the patient and provide appropriate treatment at early stages. Thus, helps in improving the standard of health care in India. The existing systems for analysis of diabetes data takes more time, inaccurate and cannot handle large amount of data. In order to overcome this drawback, automated method is proposed in this paper to handle large amount of diabetes data and to classify it as Type1 and Type2. The proposed method uses Hadoop environment coupled with Map Reduce technique to handle large amount of data. Support Vector Machine (SVM) algorithm is used for classification of diabetes into Type 1, Type 2 and Normal. The experiment is carried out on data ranging from 100 MB to 2 GB. Once the data is classified into Type 1 and Type 2, similar data can be retrieved from the hospital database. Based on this result, effective treatment can be provided to the patient.