Big Data Mining using Map Reduce: A Survey Paper

Big data is large volume, heterogeneous, distributed data. Big data applications where data collection has grown continuously, it is expensive to manage, capture or extract and process data using existing software tools. For example Weather Forecasting, Electricity Demand Supply, social media and so on. With increasing size of data in data warehouse it is expensive to perform data analysis. Data cube commonly abstracting and summarizing databases. It is way of structuring data in different n dimensions for analysis over some measure of interest. For data processing Big data processing framework relay on cluster computers and parallel execution framework provided by Map-Reduce. Extending cube computation techniques to this paradigm. MR-Cube is framework (based on mapreduce)used for cube materialization and mining over massive datasets using holistic measure. MR-Cube efficiently computes cube with holistic measures over billion-tuple datasets.