Two linked ensembles are used for a supervised learning
problem with rare-event counts. With many target instances of zero, more traditional loss functions (such as squared error and class error) are often not relevant and a statistical model leads to a likelihood with two related parameters from a zero-inflated Poisson (ZIP) distribution. In a new approach, a linked pair of gradient boosted tree ensembles are developed to handle the multiple parameters in a manner that can be generalized to other problems. The result is a unique learner that extends machine learning methods to data with nontraditional structures. We empirically compare to two real data sets and two artificial data sets versus a single-tree approach (ZIP-tree) and a statistical generalized linear model.
%0 Journal Article
%1 borisov_zero-inflated_2009
%A Borisov, Alexander
%A Runger, G.
%A Tuv, E.
%A Lurponglukana-Strand, Nutta
%D 2009
%J Advances in Intelligent Data Analysis VIII
%K Boosting, Learning, Machine Poisson, \_tablet, inflated regression trees, zero
%P 225--236
%R 10.1007/978-3-642-03915-7_20
%T Zero-inflated boosted ensembles for rare event counts
%U http://www.springerlink.com/index/5006gt8610u040u3.pdf
%V 5572
%X Two linked ensembles are used for a supervised learning
problem with rare-event counts. With many target instances of zero, more traditional loss functions (such as squared error and class error) are often not relevant and a statistical model leads to a likelihood with two related parameters from a zero-inflated Poisson (ZIP) distribution. In a new approach, a linked pair of gradient boosted tree ensembles are developed to handle the multiple parameters in a manner that can be generalized to other problems. The result is a unique learner that extends machine learning methods to data with nontraditional structures. We empirically compare to two real data sets and two artificial data sets versus a single-tree approach (ZIP-tree) and a statistical generalized linear model.
@article{borisov_zero-inflated_2009,
abstract = {Two linked ensembles are used for a supervised learning
problem with rare-event counts. With many target instances of zero, more traditional loss functions (such as squared error and class error) are often not relevant and a statistical model leads to a likelihood with two related parameters from a zero-inflated Poisson (ZIP) distribution. In a new approach, a linked pair of gradient boosted tree ensembles are developed to handle the multiple parameters in a manner that can be generalized to other problems. The result is a unique learner that extends machine learning methods to data with nontraditional structures. We empirically compare to two real data sets and two artificial data sets versus a single-tree approach (ZIP-tree) and a statistical generalized linear model.},
added-at = {2017-01-09T13:57:26.000+0100},
author = {Borisov, Alexander and Runger, G. and Tuv, E. and Lurponglukana-Strand, Nutta},
biburl = {https://www.bibsonomy.org/bibtex/23df54c8714ffe32d0a619256366bf318/yourwelcome},
doi = {10.1007/978-3-642-03915-7_20},
interhash = {5b9d7b89dd4d5faf1a67a7c54a2d6916},
intrahash = {3df54c8714ffe32d0a619256366bf318},
journal = {Advances in Intelligent Data Analysis VIII},
keywords = {Boosting, Learning, Machine Poisson, \_tablet, inflated regression trees, zero},
pages = {225--236},
timestamp = {2017-01-09T14:01:11.000+0100},
title = {Zero-inflated boosted ensembles for rare event counts},
url = {http://www.springerlink.com/index/5006gt8610u040u3.pdf},
urldate = {2012-08-04},
volume = 5572,
year = 2009
}