Ensemble classifier and resampling for imbalanced multiclass learning
An ensemble classifier called DECIML has previously reported that the classifier is able to perform on benchmark data compared to several single classifiers and ensemble classifiers such as AdaBoost, Bagging and Random Forest.The implementation of the ensemble using sampling was carried out in orde...
保存先:
| 主要な著者: | , , |
|---|---|
| フォーマット: | Conference or Workshop Item |
| 言語: | English |
| 出版事項: |
2015
|
| 主題: | |
| オンライン･アクセス: | http://repo.uum.edu.my/15678/1/PID047.pdf http://repo.uum.edu.my/15678/ http://www.icoci.cms.net.my/proceedings/2015/TOC.html |
| タグ: |
タグ追加
タグなし, このレコードへの初めてのタグを付けませんか!
|
| 要約: | An ensemble classifier called DECIML has previously reported that the classifier is able to perform on benchmark data compared to several single classifiers and ensemble classifiers such as AdaBoost, Bagging
and Random Forest.The implementation of the ensemble using sampling was carried out in order to investigate if there are any improvements in the classification performances of the DECIML.Random sampling with replacement (SWR) method is applied to minority class in the imbalanced multiclass data. Results show that the SWR is able to increase the average performance of the ensemble classifier |
|---|