Locally weighted regression (LWR) is a variation of the standard linear regression technique in which training points close to the query point have more influence over the fitted regression surface. Given a set of training points, linear regression ts the linear model that minimizes squared prediction error over
the whole training set.
source on git :https://github.com/ahrnazemi/jlwr
This implicitly assumes that we know the global form of the underlying function that generated the data.
Be the first to post a review of JLWR!