A Comparison of Ensemble Creation Techniques

Banfield, Robert E.; Hall, Lawrence O.; Bowyer, Kevin W.; Bhadoria, Divya; Kegelmeyer, W. Philip; Eschrich, Steven A.

doi:10.1007/978-3-540-25966-4_22

Cited by 36 publications

(27 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Then we used Breiman's random forest (RF) algorithm [23], with 250 unpruned trees per partition with both unweighted and weighted (RFW) predictions. Its accuracy was evaluated in [24] and shown to be comparable with or better than other well known ensemble generation techniques. The number of random features chosen at each decision tree node was log 2 n + 1 given n features.…”

Section: Classification Systemmentioning

confidence: 95%

Using classifier ensembles to label spatially disjoint data

Shoemaker¹,

Banfield²,

Hall³

et al. 2008

Information Fusion

Self Cite

View full text Add to dashboard Cite

Section: Classification Systemmentioning

confidence: 95%

Using classifier ensembles to label spatially disjoint data

Shoemaker¹,

Banfield²,

Hall³

et al. 2008

Information Fusion

Self Cite

View full text Add to dashboard Cite

“…The final classifier, additionally, usually achieves a high degree of accuracy in the test set as various authors have shown both theoretically and empirically (Banfield et al, 2004;Bauer & Kohavi, 1999;Dietterich, 2000;Drucker & Cortes, 1996;Drucker, Cortes, Jackel, LeCun, & Vapnik, 1994;Opitz & Maclin, 1999;Schapire, Freund, Bartlett, & Lee, 1997). Even though there are several versions of the boosting algorithms (Friedman, Hastie, & Tibshirani, 2000), the most widely used is the one by Freund and Schapire (1996) which is known as AdaBoost.…”

Section: Boostingmentioning

confidence: 97%

The individual borrowers recognition: Single and ensemble trees

Chrzanowska

Alfaro

Witkowska

2009

Expert Systems with Applications

View full text Add to dashboard Cite

“…Once the process has finished, the single classifiers obtained are combined into a final, highly accurate classifier in the training set. The final classifier therefore usually achieves a high degree of accuracy in the test set as various authors have shown both theoretically and empirically [5][6][7][8][9][10][11][12][13][14][15].…”

Section: Boostingmentioning

confidence: 99%

A boosting approach for corporate failure prediction

2006

View full text Add to dashboard Cite

Predicting corporate failure is an important management science problem. This is a typical classification question where the objective is to determine which indicators are involved in the failure/success of a corporation. Despite the importance of this problem, until now only classical machine learning tools have been considered to tackle this classification task. The objective of this paper is twofold. On the one hand, we introduce novel discerning measures to rank independent variables in a generic classification task. On the other hand, we apply boosting techniques to improve the accuracy of a classification tree. We apply this methodology to a set of European firms, considering the usual predicting variables such as financial ratios, as well as including novel variables rarely used before in corporate failure prediction, such as firm size, activity and legal structure. We show that our approach decreases the generalization error about thirty percent with respect to the error produced with a classification tree. In addition, the most important ratios deal with profitability and indebtedness, as is usual in failure prediction studies.

show abstract

A Comparison of Ensemble Creation Techniques

Cited by 36 publications

References 9 publications

Using classifier ensembles to label spatially disjoint data

Using classifier ensembles to label spatially disjoint data

The individual borrowers recognition: Single and ensemble trees

A boosting approach for corporate failure prediction

Contact Info

Product

Resources

About