Improved Threshold Selection by Using Calibrated Probabilities for Random Forest Classifiers

Florian Baumann; Jinghui Chen; Karsten Vogt; Bodo Rosenhahn

doi:10.1109/CRV.2015.28

Improved Threshold Selection by Using Calibrated Probabilities for Random Forest Classifiers

Florian Baumann
, Jinghui Chen
, Karsten Vogt
, Bodo Rosenhahn

College of Information Sciences and Technology

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

1 Scopus citations

All Science Journal Classification (ASJC) codes

Computer Vision and Pattern Recognition
Computer Science Applications

Cite this

Baumann, F., Chen, J., Vogt, K., & Rosenhahn, B. (2015). Improved Threshold Selection by Using Calibrated Probabilities for Random Forest Classifiers. In Proceedings -2015 12th Conference on Computer and Robot Vision, CRV 2015 (pp. 155-160). Article 7158334 (Proceedings -2015 12th Conference on Computer and Robot Vision, CRV 2015). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/CRV.2015.28

Baumann, Florian ; Chen, Jinghui ; Vogt, Karsten et al. / Improved Threshold Selection by Using Calibrated Probabilities for Random Forest Classifiers. Proceedings -2015 12th Conference on Computer and Robot Vision, CRV 2015. Institute of Electrical and Electronics Engineers Inc., 2015. pp. 155-160 (Proceedings -2015 12th Conference on Computer and Robot Vision, CRV 2015).

@inproceedings{1b0ff9826ff64ef7b8ca73a7af5657f1,

title = "Improved Threshold Selection by Using Calibrated Probabilities for Random Forest Classifiers",

abstract = "Random Forest is a well-known ensemble learning method that achieves high recognition accuracies while preserving a fast training procedure. To construct a Random Forest classifier, several decision trees are arranged in a forest while a majority voting leads to the final decision. In order to split each node of a decision tree into two children, several possible variables are randomly selected while a splitting criterion is computed for each of them. Using this pool of possible splits, the Random Forest algorithm selects the best variable according to the splitting criterion. Often, this splitting is not reliable leading to a reduced recognition accuracy. In this paper, we propose to introduce an additional condition for selecting the best variable leading to an improvement of the recognition accuracy, especially for a smaller number of trees. We enhance the standard threshold selection by a quality estimation that is computed using a calibration method. The proposed method is evaluated on datasets for machine learning as well as object recognition.",

author = "Florian Baumann and Jinghui Chen and Karsten Vogt and Bodo Rosenhahn",

note = "Publisher Copyright: {\textcopyright} 2015 IEEE.; 12th Conference on Computer and Robot Vision, CRV 2015 ; Conference date: 03-06-2015 Through 05-06-2015",

year = "2015",

month = jul,

day = "14",

doi = "10.1109/CRV.2015.28",

language = "English (US)",

series = "Proceedings -2015 12th Conference on Computer and Robot Vision, CRV 2015",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "155--160",

booktitle = "Proceedings -2015 12th Conference on Computer and Robot Vision, CRV 2015",

address = "United States",

}

Improved Threshold Selection by Using Calibrated Probabilities for Random Forest Classifiers. / Baumann, Florian; Chen, Jinghui; Vogt, Karsten et al.
Proceedings -2015 12th Conference on Computer and Robot Vision, CRV 2015. Institute of Electrical and Electronics Engineers Inc., 2015. p. 155-160 7158334 (Proceedings -2015 12th Conference on Computer and Robot Vision, CRV 2015).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Improved Threshold Selection by Using Calibrated Probabilities for Random Forest Classifiers

AU - Baumann, Florian

AU - Chen, Jinghui

AU - Vogt, Karsten

AU - Rosenhahn, Bodo

PY - 2015/7/14

Y1 - 2015/7/14

N2 - Random Forest is a well-known ensemble learning method that achieves high recognition accuracies while preserving a fast training procedure. To construct a Random Forest classifier, several decision trees are arranged in a forest while a majority voting leads to the final decision. In order to split each node of a decision tree into two children, several possible variables are randomly selected while a splitting criterion is computed for each of them. Using this pool of possible splits, the Random Forest algorithm selects the best variable according to the splitting criterion. Often, this splitting is not reliable leading to a reduced recognition accuracy. In this paper, we propose to introduce an additional condition for selecting the best variable leading to an improvement of the recognition accuracy, especially for a smaller number of trees. We enhance the standard threshold selection by a quality estimation that is computed using a calibration method. The proposed method is evaluated on datasets for machine learning as well as object recognition.

AB - Random Forest is a well-known ensemble learning method that achieves high recognition accuracies while preserving a fast training procedure. To construct a Random Forest classifier, several decision trees are arranged in a forest while a majority voting leads to the final decision. In order to split each node of a decision tree into two children, several possible variables are randomly selected while a splitting criterion is computed for each of them. Using this pool of possible splits, the Random Forest algorithm selects the best variable according to the splitting criterion. Often, this splitting is not reliable leading to a reduced recognition accuracy. In this paper, we propose to introduce an additional condition for selecting the best variable leading to an improvement of the recognition accuracy, especially for a smaller number of trees. We enhance the standard threshold selection by a quality estimation that is computed using a calibration method. The proposed method is evaluated on datasets for machine learning as well as object recognition.

UR - https://www.scopus.com/pages/publications/84943159205

UR - https://www.scopus.com/pages/publications/84943159205#tab=citedBy

U2 - 10.1109/CRV.2015.28

DO - 10.1109/CRV.2015.28

M3 - Conference contribution

AN - SCOPUS:84943159205

T3 - Proceedings -2015 12th Conference on Computer and Robot Vision, CRV 2015

SP - 155

EP - 160

BT - Proceedings -2015 12th Conference on Computer and Robot Vision, CRV 2015

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 12th Conference on Computer and Robot Vision, CRV 2015

Y2 - 3 June 2015 through 5 June 2015

ER -