|“||the discovery of meaningful patterns in data, and is one of the steps in the data life cycle of collection of raw data, preparation of information, analysis of patterns to synthesize knowledge, and action to produce value.||”|
"Analytic processes are often characterized as discovery for the initial hypothesis formulation, development for establishing the analytics process for a specific hypothesis, and applied for the encapsulation of the analysis into an operational system. While Big Data has touched all three types of analytic processes, the majority of the changes is observed in development and applied analytics. New Big Data engineering technologies change the types of analytics that are possible, but do not result in completely new types of analytics. However, given the retrieval speeds, analysts are able to interact with their data in ways that were not previously possible. Traditional statistical analytic techniques downsize, sample, or summarize the data before analysis. This was done to make analysis of large datasets reasonable on hardware that could not scale to the size of the dataset. Big Data analytics often emphasize the value of computation across the entire dataset, which gives analysts better chances to determine causation, rather than just correlation. Correlation, though, is still useful when knowing the direction or trend of something is enough to take action. Today, most analytics in statistics and data mining focus on causation—being able to describe why something is happening. Discovering the cause aids actors in changing a trend or outcome."
- NIST Big Data Interoperability Framework, Vol. 1, at 8.
- Id. at 15.