Abstract | ||
---|---|---|
It has been recognized that Classification trees (CART) are unstable; a small perturbation in the input variables or a fresh sample can lead to a very different classification tree. Some approaches exist that try to correct this instability. However, their benefits can, at present, be appreciated only qualitatively. A similarity measure between two classification trees is introduced that can measure their closeness. Its usefulness is illustrated with synthetic data on the impact of radioactivity deposit through the environment. In this context, a modified node level stabilizing technique, referred to as the NLS-REP method, is introduced and shown to be more stable than the classical CART method. |
Year | DOI | Venue |
---|---|---|
2009 | 10.1016/j.csda.2008.10.033 | Computational Statistics & Data Analysis |
Keywords | Field | DocType |
small perturbation,input variable,classical cart method,modified node level,fresh sample,radioactivity deposit,similarity measure,different classification tree,classification tree,nls-rep method,synthetic data | Econometrics,Similarity measure,Closeness,Cart,Synthetic data,Linear discriminant analysis,Statistics,Multivariate analysis,Decision tree learning,Numerical stability,Mathematics | Journal |
Volume | Issue | ISSN |
53 | 4 | Computational Statistics and Data Analysis |
Citations | PageRank | References |
10 | 0.63 | 10 |
Authors | ||
4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Bénédicte Briand | 1 | 10 | 0.63 |
Gilles R. Ducharme | 2 | 16 | 4.02 |
Vanessa Parache | 3 | 10 | 0.63 |
Catherine Mercat-Rommens | 4 | 10 | 0.63 |