Thursday, June 25, 2009

Automatically Learning From Data - Logistic Regression With L2 Regularization in Python

Suppose you have data describing a bunch of buildings and earthquakes (E.g., year the building was constructed, type of material used, the strength of earthquake, etc), and you know whether each building collapsed or not in each past earthquake. Using this data, you'd like to make predictions about whether a given building is going to collapse in a hypothetical future earthquake -- you can see how this would be useful to say, insurance companies. Read more ...