BIGDATA and statistics
First, what is big data?
Big data ,which was generated by the advance of technology of IT and the spread of the Internet, is generic name of a new mechanism which deal with the more variety data than before.
Big data also has four features.
The feature of big data is huge volume. There are a lot of data inside corporation and outside corporation. Big data is from terabytes to petabytes. The more data become, the more Amount of calculation become.
Big data is not necessarily the structured data like dealing with enterprise systems. There are no structured data; Text, Voice, Video, Clickstream and Log file and so on. The movement which use these data in bussiness spread around the world.
The data is frequently generated from sensor and IC tag now. Market environment in these days request to the corresponding in real time by these data.
Inconsistency of data, Uncertainty by ambiguity, Inaccuracies stacked approximation are rejected, and it is important to do the mind by reliable data.
But the statistics become approximation if based data is very very big.
For example, the statistics is used in audience survey of television. If the purpose is clear even, it is not necessary that there are huge data. It is the statistics.
In short, the essence of statistics deny the thought of big data. So it is possible to borrow the technic from these area, it is impossible to borrow fundamental thought of it from that.