Data Set Balancing Export

@incollection{citeulike:5700900, abstract = {This paper conducts experiments with three skewed data sets, seeking to demonstrate problems when skewed data is used, and identifying counter problems when data is balanced. The basic data mining algorithms of decision tree, regression-based, and neural network models are considered, using both categorical and continuous data. Two of the data sets have binary outcomes, while the third has a set of four possible outcomes. Key findings are that when the data is highly unbalanced, algorithms tend to degenerate by assigning all cases to the most common outcome. When data is balanced, accuracy rates tend to decline. If data is balanced, that reduces the training set size, and can lead to the degeneracy of model failure through omission of cases encountered in the test set. Decision tree algorithms were found to be the most robust with respect to the degree of balancing applied.}, author = {Olson, David L.}, citeulike-article-id = {5700900}, citeulike-linkout-0 = {http://www.springerlink.com/content/xpvb18c47u4lvug7}, journal = {Data Mining and Knowledge Management}, keywords = {machine\_learning}, pages = {71--80}, posted-at = {2009-09-01 21:35:39}, priority = {2}, title = {Data Set Balancing}, url = {http://www.springerlink.com/content/xpvb18c47u4lvug7}, year = {2005} }

TY - CHAP ID - citeulike:5700900 L3 - citeulike-article-id:5700900 N2 - This paper conducts experiments with three skewed data sets, seeking to demonstrate problems when skewed data is used, and identifying counter problems when data is balanced. The basic data mining algorithms of decision tree, regression-based, and neural network models are considered, using both categorical and continuous data. Two of the data sets have binary outcomes, while the third has a set of four possible outcomes. Key findings are that when the data is highly unbalanced, algorithms tend to degenerate by assigning all cases to the most common outcome. When data is balanced, accuracy rates tend to decline. If data is balanced, that reduces the training set size, and can lead to the degeneracy of model failure through omission of cases encountered in the test set. Decision tree algorithms were found to be the most robust with respect to the degree of balancing applied. JF - Data Mining and Knowledge Management EP - 80 TI - Data Set Balancing SP - 71 KW - machine_learning AU - Olson, David PY - 2005/// UR - http://www.springerlink.com/content/xpvb18c47u4lvug7 ER -