Big Data

The term "big data" refers to voluminous, rapidly growing, complicated datasets; and not specifically to a particular size of the data set. The importance of big data lies in the information which can be obtained from it through predictive analytics. For big data analytics, we require tools and algorithms which can scale up to data volume with high performance, while keeping other parameters of data like variety, etc. also under consideration. The traditional data mining tools and algorithms are not able to handle such large fast-growing datasets, so we need new tools and algorithms to handle large volumes of complex data. In this paper, the authors propose a scalable architecture for warehousing and mining of structured and unstructured datasets.