Big data plays an important role in healthcare. With an increase in growth of data, processing and analyzing them becomes a challenging task. Diagnosis and prediction of disease becomes difficult especially when it comes to Big Data. Clustering is one of the Data Mining tools that help us to analyze Big Data effectively. Existing algorithms have high computational complexity and they do not perform well especially when it comes to Big Data. Since most of the data is unstructured a graph based spectral technique using power method is chosen for analysis. The algorithm is made more effective by making them to converge using extrapolation technique. Moreover, they are designed to handle larger datasets using MapReduce framework. The main objective of this paper is to help the doctors in predicting the diseases more accurately using the proposed algorithm. Experiments were conducted on various synthetic datasets and real data’s to prove the algorithmic efficiency and accuracy.
CITATION STYLE
Jayalatchumy, D., & Thambidurai, P. (2017). Prediction of diseases using Hadoop in big data - A modified approach. In Advances in Intelligent Systems and Computing (Vol. 573, pp. 229–238). Springer Verlag. https://doi.org/10.1007/978-3-319-57261-1_23
Mendeley helps you to discover research relevant for your work.