Tutorial: Getting Started with MART in R

intended to be a tutorial introduction. Minimal knowledge concerning the technical details of the MART methodology or the use of the R statistical package …

More PDF Content

Abstract Multiple additive regression trees (MART) is a methodology for predictive data mining (regression and classification). This note illustrates the use of the R/MART interface. It is intended to be a tutorial introduction. Minimal knowledge concerning the technical details of the MART methodology or the use of the R statistical package is presumed.
Predictive data mining is concerned with constructing statistical models from historical data. These models are used to predict future unknown data values, and/or to help gain an understanding of the predictive relationships represented in the data. The data consists of sets of measurements made on objects (observations) collected in a data base. One of the measured variables (denoted by y ) is designated as the one to be predicted, given future values of the other variables denoted by x = {x1 , x2 , * * *, xn }. Depending on the field of study, y is referred to as the response variable (Statistics), output variable (neural nets), or concept (Machine Learning). The x-variables are referred to as predictor variables, input variables, or attributes in these respective fields. The data base consists of a collection of N previously solved cases {yi , xi1 , * * *, xin }i=1 . The predictive model takes the abstract form  y = f (x1 , * * *, xn ),  (2)

Download Tutorial: Getting Started with MART in R pdf from www-stat.stanford.edu, 24 pages, 243.89KB.
Related Books

Leave a Reply