Smart Healthcare System for disease administration and prediction

The rapid growth of population has paved way for the vast spread of healthcare. Huge amount of data are available from which predictions can be attempted. Contemporary approaches predict a patient's disease based on the disease rules generated by decision tree algorithms. The number of attributes considered for predicting a disease varies based on algorithms. This paper applies Amoeba algorithm for generating decision rules and rules are stored in Hbase. Amoeba algorithm does not require confidence and support for finding frequent itemsets. The attributes considered for rules generation in Amoeba algorithm are also less compared to Apriori and Fp-growth algortihms. So efficient processing is guaranteed. Basically, data are available in structured, semi-structured and unstructured form. Text mining is employed to convert different forms of datato structured data and store in Hbase. The patient can predict a disease and know the risk of the disease further by comparing the disease rules with patient's current status and family history. This paper proposes a Smart Healthcare System(SHS) using Hadoop with Text Mining which can efficiently predict disease based on the data available.