Please vote on www.kdnuggets.com - see explanation below
Physicist Eugene Wigner wrote "The Unreasonable Effectiveness of Mathematics in the Natural Sciences" where he observed that many physical phenomena are governed by laws which can be concisely stated in mathematical terms.
Similar approaches to find concise mathematical laws were not as successful for most problems outside of physics.
In fact, a recent paper The Unreasonable Effectiveness of Data, IEEE Intelligent Systems, 2009, by Alon Halevy, Peter Norvig, Fernando Pereira, observed that
Problems that involve interacting with humans, such as natural language understanding, have not proven to be solvable by concise, neat formulas like F = ma. Instead, the best approach appears to be to embrace the complexity of the domain and address it by harnessing the power of data
However, the danger of embracing purely data-driven approach is that data miners may ignore potentially useful concise patterns which may exist in data, and even if they will not have physics-level accuracy, they may still be very useful.
What is your opinion:
Are there concise, mathematical laws for large datasets in business, social, and biology data?
Please vote on www.kdnuggets.com
See also KDnuggets 10:n04 Quote
Had computers been invented in Copernicus time, then rather than discovering how everything revolves around the sun, he might have invented elaborate epicycles that explained how everything revolves around the Earth - and we still might be using them.David Harrison, Jackson Laboratory gerontologist, on the tendency to accumulate and analyze vast datasets rather than searching for underlying laws. (Thanks to Tom Fawcett)