Neighbourhood systems based knowledge acquisition using MapReduce from Big Data over cloud computing

Cloud computing represents a paradigm shift - a transition from computing-as-a-product to computing-as-aservice. It can be applied to a range of areas, including ecommerce, health, education, communities, etc which are emerging as the important sectors in today's market. Day-byday more knowledge is added to the internet and is shared amongst the users in addition to the official use over the cloud. As a result the energy consumption by the networks is increasing and it needs to be managed. This usage can be brought into account for measuring and hence conserving the energy. The consumption is all together considered for the processing, storage and transport of the knowledge granules over the cloud. Since the data accessed in the cloud is “ondemand”, the prediction techniques like those using rough sets can be used to minimize the transfer of data over the cloud networks. The data over the cloud can be procured with the help of rough set based methods efficiently which can help in conserving the energy. Recently, the basic rough set theory has been extended to the notion of neighbourhood based rough sets where information systems with heterogeneous features are prevalent. In this paper, we propose a neighbourhood based rough set approach for knowledge acquisition using MapReduce from Big Data.