13 articles
Articles prepublished February 07, 2012
 |
|
An Experimental Evaluation of Boosting Methods for Classification
Original Article
R. Stollhoff (1), W. Sauerbrei (2), M. Schumacher (2)
(1) Max-Planck Institute for Mathematics in the Sciences, Leipzig, Germany; (2) Institute of Medical Biometry and Medical Informatics, University Medical Center, Freiburg, Germany
Summary
Objectives: In clinical medicine, the accuracy achieved by classification rules is often not sufficient to justify their use in daily practice. In order to improve classifiers it has become popular to combine single classification rules into a classification ensemble. Two popular boosting methods will be compared with classical statistical approaches. Methods: Using data from a clinical study on the diagnosis of breast tumors and by simulation we will compare AdaBoost with gradient boosting ensembles of regression trees. We will also consider a tree approach and logistic regression as traditional competitors. In logistic regression we allow to select non- linear effects by the fractional polynomial approach. Performance of the classifiers will be assessed by estimated misclassification rates and the Brier score. Results: We will show that boosting of simple base classifiers gives classification rules with improved predictive ability. However, the performance of boosting classifiers was not generally superior to the performance of logistic regression. In contrast to the computer-intensive methods the latter are based on classifiers which are much easier to interpret and to use. Conclusions: In medical applications, the logistic regression model remains a method of choice or, at least, a serious competitor of more sophisticated techniques. Refinement of boosting methods by using optimized number of boosting steps may lead to further improvement.
Keywords
Classification, simulation study, boosting, generalized additive models, diagnosis of breast tumors
DOI
http://dx.doi.org/10.3414/ME0543You may also be interested in...
| 1. | ||
W. Adler, A. Peters, B. Lausen Methods of Information in Medicine 2008 47 1: 38-46 http://dx.doi.org/10.3414/ME0348 | ||
| 2. | ||
J. E. Mezzich1, I. M. Salloum2 Die Psychiatrie 2007 4 4: 262-265 | ||
| 3. | ||
A. Doering (1) , H. Jäger (1) , H. Witte (1) , M. Galicki (1) , C. Schelenz (2) , M. Specht (2) , K. Reinhart (2) , M. Eiselt (3) Methods of Information in Medicine 1999 38 3: 214-224 | ||
Preprint Online November 21, 2011
|
|
An Experimental Evaluation of Boosting Methods for Classification
Original Article
R. Stollhoff (1), W. Sauerbrei (2), M. Schumacher (2)
(1) Max-Planck Institute for Mathematics in the Sciences, Leipzig, Germany; (2) Institute of Medical Biometry and Medical Informatics, University Medical Center, Freiburg, Germany
Summary
Objectives: In clinical medicine, the accuracy achieved by classification rules is often not sufficient to justify their use in daily practice. In order to improve classifiers it has become popular to combine single classification rules into a classification ensemble. Two popular boosting methods will be compared with classical statistical approaches. Methods: Using data from a clinical study on the diagnosis of breast tumors and by simulation we will compare AdaBoost with gradient boosting ensembles of regression trees. We will also consider a tree approach and logistic regression as traditional competitors. In logistic regression we allow to select non- linear effects by the fractional polynomial approach. Performance of the classifiers will be assessed by estimated misclassification rates and the Brier score. Results: We will show that boosting of simple base classifiers gives classification rules with improved predictive ability. However, the performance of boosting classifiers was not generally superior to the performance of logistic regression. In contrast to the computer-intensive methods the latter are based on classifiers which are much easier to interpret and to use. Conclusions: In medical applications, the logistic regression model remains a method of choice or, at least, a serious competitor of more sophisticated techniques. Refinement of boosting methods by using optimized number of boosting steps may lead to further improvement.
Keywords
Classification, simulation study, boosting, generalized additive models, diagnosis of breast tumors
DOI
http://dx.doi.org/10.3414/ME0543You may also be interested in...
| 1. | ||
W. Adler, A. Peters, B. Lausen Methods of Information in Medicine 2008 47 1: 38-46 http://dx.doi.org/10.3414/ME0348 | ||
| 2. | ||
J. E. Mezzich1, I. M. Salloum2 Die Psychiatrie 2007 4 4: 262-265 | ||
| 3. | ||
A. Doering (1) , H. Jäger (1) , H. Witte (1) , M. Galicki (1) , C. Schelenz (2) , M. Specht (2) , K. Reinhart (2) , M. Eiselt (3) Methods of Information in Medicine 1999 38 3: 214-224 | ||
Articles prepublished September 14, 2010
|
|
An Experimental Evaluation of Boosting Methods for Classification
Original Article
R. Stollhoff (1), W. Sauerbrei (2), M. Schumacher (2)
(1) Max-Planck Institute for Mathematics in the Sciences, Leipzig, Germany; (2) Institute of Medical Biometry and Medical Informatics, University Medical Center, Freiburg, Germany
Summary
Objectives: In clinical medicine, the accuracy achieved by classification rules is often not sufficient to justify their use in daily practice. In order to improve classifiers it has become popular to combine single classification rules into a classification ensemble. Two popular boosting methods will be compared with classical statistical approaches. Methods: Using data from a clinical study on the diagnosis of breast tumors and by simulation we will compare AdaBoost with gradient boosting ensembles of regression trees. We will also consider a tree approach and logistic regression as traditional competitors. In logistic regression we allow to select non- linear effects by the fractional polynomial approach. Performance of the classifiers will be assessed by estimated misclassification rates and the Brier score. Results: We will show that boosting of simple base classifiers gives classification rules with improved predictive ability. However, the performance of boosting classifiers was not generally superior to the performance of logistic regression. In contrast to the computer-intensive methods the latter are based on classifiers which are much easier to interpret and to use. Conclusions: In medical applications, the logistic regression model remains a method of choice or, at least, a serious competitor of more sophisticated techniques. Refinement of boosting methods by using optimized number of boosting steps may lead to further improvement.
Keywords
Classification, simulation study, boosting, generalized additive models, diagnosis of breast tumors
DOI
http://dx.doi.org/10.3414/ME0543You may also be interested in...
| 1. | ||
W. Adler, A. Peters, B. Lausen Methods of Information in Medicine 2008 47 1: 38-46 http://dx.doi.org/10.3414/ME0348 | ||
| 2. | ||
J. E. Mezzich1, I. M. Salloum2 Die Psychiatrie 2007 4 4: 262-265 | ||
| 3. | ||
A. Doering (1) , H. Jäger (1) , H. Witte (1) , M. Galicki (1) , C. Schelenz (2) , M. Specht (2) , K. Reinhart (2) , M. Eiselt (3) Methods of Information in Medicine 1999 38 3: 214-224 | ||
Preprint Online August 05, 2011
|
|
An Experimental Evaluation of Boosting Methods for Classification
Original Article
R. Stollhoff (1), W. Sauerbrei (2), M. Schumacher (2)
(1) Max-Planck Institute for Mathematics in the Sciences, Leipzig, Germany; (2) Institute of Medical Biometry and Medical Informatics, University Medical Center, Freiburg, Germany
Summary
Objectives: In clinical medicine, the accuracy achieved by classification rules is often not sufficient to justify their use in daily practice. In order to improve classifiers it has become popular to combine single classification rules into a classification ensemble. Two popular boosting methods will be compared with classical statistical approaches. Methods: Using data from a clinical study on the diagnosis of breast tumors and by simulation we will compare AdaBoost with gradient boosting ensembles of regression trees. We will also consider a tree approach and logistic regression as traditional competitors. In logistic regression we allow to select non- linear effects by the fractional polynomial approach. Performance of the classifiers will be assessed by estimated misclassification rates and the Brier score. Results: We will show that boosting of simple base classifiers gives classification rules with improved predictive ability. However, the performance of boosting classifiers was not generally superior to the performance of logistic regression. In contrast to the computer-intensive methods the latter are based on classifiers which are much easier to interpret and to use. Conclusions: In medical applications, the logistic regression model remains a method of choice or, at least, a serious competitor of more sophisticated techniques. Refinement of boosting methods by using optimized number of boosting steps may lead to further improvement.
Keywords
Classification, simulation study, boosting, generalized additive models, diagnosis of breast tumors
DOI
http://dx.doi.org/10.3414/ME0543You may also be interested in...
| 1. | ||
W. Adler, A. Peters, B. Lausen Methods of Information in Medicine 2008 47 1: 38-46 http://dx.doi.org/10.3414/ME0348 | ||
| 2. | ||
J. E. Mezzich1, I. M. Salloum2 Die Psychiatrie 2007 4 4: 262-265 | ||
| 3. | ||
A. Doering (1) , H. Jäger (1) , H. Witte (1) , M. Galicki (1) , C. Schelenz (2) , M. Specht (2) , K. Reinhart (2) , M. Eiselt (3) Methods of Information in Medicine 1999 38 3: 214-224 | ||
Preprint Online July 26, 2011
|
|
An Experimental Evaluation of Boosting Methods for Classification
Original Article
R. Stollhoff (1), W. Sauerbrei (2), M. Schumacher (2)
(1) Max-Planck Institute for Mathematics in the Sciences, Leipzig, Germany; (2) Institute of Medical Biometry and Medical Informatics, University Medical Center, Freiburg, Germany
Summary
Objectives: In clinical medicine, the accuracy achieved by classification rules is often not sufficient to justify their use in daily practice. In order to improve classifiers it has become popular to combine single classification rules into a classification ensemble. Two popular boosting methods will be compared with classical statistical approaches. Methods: Using data from a clinical study on the diagnosis of breast tumors and by simulation we will compare AdaBoost with gradient boosting ensembles of regression trees. We will also consider a tree approach and logistic regression as traditional competitors. In logistic regression we allow to select non- linear effects by the fractional polynomial approach. Performance of the classifiers will be assessed by estimated misclassification rates and the Brier score. Results: We will show that boosting of simple base classifiers gives classification rules with improved predictive ability. However, the performance of boosting classifiers was not generally superior to the performance of logistic regression. In contrast to the computer-intensive methods the latter are based on classifiers which are much easier to interpret and to use. Conclusions: In medical applications, the logistic regression model remains a method of choice or, at least, a serious competitor of more sophisticated techniques. Refinement of boosting methods by using optimized number of boosting steps may lead to further improvement.
Keywords
Classification, simulation study, boosting, generalized additive models, diagnosis of breast tumors
DOI
http://dx.doi.org/10.3414/ME0543You may also be interested in...
| 1. | ||
W. Adler, A. Peters, B. Lausen Methods of Information in Medicine 2008 47 1: 38-46 http://dx.doi.org/10.3414/ME0348 | ||
| 2. | ||
J. E. Mezzich1, I. M. Salloum2 Die Psychiatrie 2007 4 4: 262-265 | ||
| 3. | ||
A. Doering (1) , H. Jäger (1) , H. Witte (1) , M. Galicki (1) , C. Schelenz (2) , M. Specht (2) , K. Reinhart (2) , M. Eiselt (3) Methods of Information in Medicine 1999 38 3: 214-224 | ||
Preprint Online March 21, 2011
|
|
An Experimental Evaluation of Boosting Methods for Classification
Original Article
R. Stollhoff (1), W. Sauerbrei (2), M. Schumacher (2)
(1) Max-Planck Institute for Mathematics in the Sciences, Leipzig, Germany; (2) Institute of Medical Biometry and Medical Informatics, University Medical Center, Freiburg, Germany
Summary
Objectives: In clinical medicine, the accuracy achieved by classification rules is often not sufficient to justify their use in daily practice. In order to improve classifiers it has become popular to combine single classification rules into a classification ensemble. Two popular boosting methods will be compared with classical statistical approaches. Methods: Using data from a clinical study on the diagnosis of breast tumors and by simulation we will compare AdaBoost with gradient boosting ensembles of regression trees. We will also consider a tree approach and logistic regression as traditional competitors. In logistic regression we allow to select non- linear effects by the fractional polynomial approach. Performance of the classifiers will be assessed by estimated misclassification rates and the Brier score. Results: We will show that boosting of simple base classifiers gives classification rules with improved predictive ability. However, the performance of boosting classifiers was not generally superior to the performance of logistic regression. In contrast to the computer-intensive methods the latter are based on classifiers which are much easier to interpret and to use. Conclusions: In medical applications, the logistic regression model remains a method of choice or, at least, a serious competitor of more sophisticated techniques. Refinement of boosting methods by using optimized number of boosting steps may lead to further improvement.
Keywords
Classification, simulation study, boosting, generalized additive models, diagnosis of breast tumors
DOI
http://dx.doi.org/10.3414/ME0543You may also be interested in...
| 1. | ||
W. Adler, A. Peters, B. Lausen Methods of Information in Medicine 2008 47 1: 38-46 http://dx.doi.org/10.3414/ME0348 | ||
| 2. | ||
J. E. Mezzich1, I. M. Salloum2 Die Psychiatrie 2007 4 4: 262-265 | ||
| 3. | ||
A. Doering (1) , H. Jäger (1) , H. Witte (1) , M. Galicki (1) , C. Schelenz (2) , M. Specht (2) , K. Reinhart (2) , M. Eiselt (3) Methods of Information in Medicine 1999 38 3: 214-224 | ||
Preprint Online March 04, 2011
|
|
An Experimental Evaluation of Boosting Methods for Classification
Original Article
R. Stollhoff (1), W. Sauerbrei (2), M. Schumacher (2)
(1) Max-Planck Institute for Mathematics in the Sciences, Leipzig, Germany; (2) Institute of Medical Biometry and Medical Informatics, University Medical Center, Freiburg, Germany
Summary
Objectives: In clinical medicine, the accuracy achieved by classification rules is often not sufficient to justify their use in daily practice. In order to improve classifiers it has become popular to combine single classification rules into a classification ensemble. Two popular boosting methods will be compared with classical statistical approaches. Methods: Using data from a clinical study on the diagnosis of breast tumors and by simulation we will compare AdaBoost with gradient boosting ensembles of regression trees. We will also consider a tree approach and logistic regression as traditional competitors. In logistic regression we allow to select non- linear effects by the fractional polynomial approach. Performance of the classifiers will be assessed by estimated misclassification rates and the Brier score. Results: We will show that boosting of simple base classifiers gives classification rules with improved predictive ability. However, the performance of boosting classifiers was not generally superior to the performance of logistic regression. In contrast to the computer-intensive methods the latter are based on classifiers which are much easier to interpret and to use. Conclusions: In medical applications, the logistic regression model remains a method of choice or, at least, a serious competitor of more sophisticated techniques. Refinement of boosting methods by using optimized number of boosting steps may lead to further improvement.
Keywords
Classification, simulation study, boosting, generalized additive models, diagnosis of breast tumors
DOI
http://dx.doi.org/10.3414/ME0543You may also be interested in...
| 1. | ||
W. Adler, A. Peters, B. Lausen Methods of Information in Medicine 2008 47 1: 38-46 http://dx.doi.org/10.3414/ME0348 | ||
| 2. | ||
J. E. Mezzich1, I. M. Salloum2 Die Psychiatrie 2007 4 4: 262-265 | ||
| 3. | ||
A. Doering (1) , H. Jäger (1) , H. Witte (1) , M. Galicki (1) , C. Schelenz (2) , M. Specht (2) , K. Reinhart (2) , M. Eiselt (3) Methods of Information in Medicine 1999 38 3: 214-224 | ||
Articles
You've 177 Article(s) in your Basket.
Methods News
5 Fachzeitschriften feiern 2012 "runde" Jubiläen
2012 feiern gleich fünf der 21 Fachzeitschriften "runde" Jubiläen, die im Schattauer...
Methods issue 1/2012
Monitoring and analyzing physiological parameters under different conditions of health and disease...
Methods issue 6/2011
This issue of Methods of Information in Medicine celebrates the journal‘s first 50 years. As the...
Call for Papers
We are inviting submissions for a Focus Theme in Methods of Information in Medicine on "Web...
Invitation to Apply for the Student Editorial Board 2012/2013
The Journal Methods of Information in Medicine is now accepting applications from trainees who...
Methods issue 5/2011
In 2008 it was decided that Methods of Information in Medicine will continue the tradition of...