Seb Schmoller points to this article by three Google researchers. First, read Eugene Wigner's article The Unreasonable Effectiveness of Mathematics in the Natural Sciences to set the stage. Then follow the data. I note, specifically, the observation that the process of constructing a theory, then interpreting the data according to parameters set by the theory, is an ineffective way to proceed. Observe the data itself. "Represent all the data with a nonparametric model rather than trying to summarize it with a parametric model, because with very large data sources, the data holds a lot of detail. For natural language applications, trust that human language has already evolved words for the important concepts. See how far you can go by tying together the words that are already there, rather than by inventing new concepts with clusters of words."
Today: 3 Total: 97 [Direct link] [Share]
