.png){kind=icon}
返回
An ensemble pruning method considering classifiers' interaction based on information theory for facial expression recognition
.png){kind=icon}
PRE
AI
DOI：10.1007/s00530-023-01227-2.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Ensemble learning combines all generated base learners for better generalization performance, but weak and redundant classifiers reduce the classification system's performance, so the researchers proposed ensemble pruning. Existing ensemble pruning methods usually utilize the classifiers' diversity and ability to select optimal classifier sequences. However, they ignore that the interaction between two weak or redundant classifiers can improve performance. In this paper, we focus on the interaction between classifiers and propose a new ensemble pruning method CCIEP (An Ensemble Pruning Method Considering Classifiers' Interaction). We apply CCIEP mainly to facial expression recognition. CCIEP consists of two parts: first, we use symmetric uncertainty as the ranking metric, and then we use symmetric uncertainty to perform ranking-based pruning of the classifier pool to ensure the ensemble performance of a subset of classifiers. Second, we use the interaction information to explore the binding relationships between classifiers. Then, we add classifiers with interactions to the selected subset of classifiers, which can optimize the system performance by adding interacting classifiers. Experimental results demonstrate that the method outperforms some state-of-the-art ensemble pruning methods on five classical face expression data sets and 10 UCI data sets.
Keyword:
Ensemble pruning
Classifiers' interaction
Symmetric uncertainty
Facial expression recognition
Ensemble learning
期刊
IF:
3.1
论文数:
2.8K
被引数:
2.7K
机构
引用论文
A novel ensemble pruning algorithm based on randomized greedy selective strategy and ballot.png)
NEUROCOMPUTING
NEUROCOMPUTINGIF6.5
PREAI
分享收藏Deep Learning and Machine Learning with Grid Search to Predict Later Occurrence of Breast Cancer Metastasis Using Clinical Data使用网格搜索的深度学习和机器学习，使用临床数据预测乳腺癌转移的发生
OAAI
OAAI分享收藏Chromosome numbers in the rare endemicPennantia baylisiana(W.R.B. Oliv.) G.T.S. Baylis (Icacinaceae) and related species稀有特有物种Pennantia baylisiana(W.R.B. Oliv.) G.T.S. Baylis（属ICACINACEAE科）及其近缘种的染色体数目
PREAI
分享收藏
查看更多内容