Optimized decision tree based models
作者： Tarun Agarwal , Robert Matthias Steele , Leo Parker Dirac , Jun Qian
DOI:
关键词:
摘要: During a training phase of machine learning model, representations at least some nodes decision tree are generated and stored on persistent storage in depth-first order. A respective predictive utility metric (PUM) value is determined for one or more nodes, indicating expected contributions the to prediction model. particular node selected removal from based partly its PUM value. modified version tree, with removed, obtaining prediction.
lens.org 参考文章(61)
Ioan Bogdan Crivat, C. James MacLennan, Machine learning semantic model ,(2013)
Wei-hao Lin, Travis H.K. Green, Robert Kaplow, Gang Fu, Gideon S. Mann, Predictive Analytical Modeling Accuracy Assessment ,(2011)
Dave A. Kennon, Gregory A. Johnson, John Perlis, Lorraine J. Webster, Decision service method and system ,(2001)
Yan Zhang, Wei Qu, System and method for autonomic tuning the number of threads in application server at runtime ,(2007)
Reinhard Sebastian Bernhard Nowozin, Decision tree training in machine learning ,(2012)
Janne Suontausta, Jilei Tian, Low memory decision tree ,(2004)
Joseph A. Coha, Timothy C. O'Konski, Ashish Karkare, Method for optimization of memory usage for a computer program ,(2002)
Clifford Behrens, Hiralal Agrawal, Balakrishnan Dasarathy, Learning program behavior for anomaly detection ,(2010)
Mohamed Zait, Khaled Yagoub, Benoit Dageville, Mohamed Ziauddin, Dinesh Das, Automatic learning optimizer ,(2004)
Stephane Chiocchetti, George H. Forman, Preparing data for machine learning ,(2004)