2,866 Hits in 6.8 sec

Addressing Class Overlap under Imbalanced Distribution: An Improved Method and Two Metrics

Zhuang Li, Jingyan Qin, Xiaotong Zhang, Yadong Wan
2021 Symmetry  
eight algorithms can be improved by 34.7488% in average.  ...  The experimental results demonstrate that the proposed class overlap metrics outperform other compared metrics for the imbalanced datasets and the Pearson correlation coefficient with the AUC metric of  ...  Therefore, both correlation coefficients with the classification performances and the generalization ability for different classification algorithms need to be improved.  ... 
doi:10.3390/sym13091649 fatcat:bm6cso3larcuvo5rbh23alcueq

Adaptive Fusion Based Method for Imbalanced Data Classification

Zefeng Liang, Huan Wang, Kaixiang Yang, Yifan Shi
2022 Frontiers in Neurorobotics  
To deal with the imbalance problem, various ensemble algorithms are proposed.  ...  However, conventional ensemble algorithms do not consider exploring an effective feature space to further improve the performance.  ...  Compared with other algorithms, the proposed algorithm has at least a 1% improvement in average AUC.  ... 
doi:10.3389/fnbot.2022.827913 pmid:35295673 pmcid:PMC8918481 fatcat:3e4gzjsjrbh4jdsdu6h3d3cq4q

Posterior Re-calibration for Imbalanced Datasets [article]

Junjiao Tian, Yen-Cheng Liu, Nathan Glaser, Yen-Chang Hsu, Zsolt Kira
2020 arXiv   pre-print
Our results on six different datasets and five different architectures show state of art accuracy, including on large-scale imbalanced datasets such as iNaturalist for classification and Synthia for semantic  ...  The resulting algorithm can be conveniently used on probabilistic classification problems agnostic to underlying architectures.  ...  This method can often improve the performance for imbalanced classification [11] [12] .  ... 
arXiv:2010.11820v1 fatcat:boabo72oqzbkpac2qm6nd3vo4y

Building Decision Tree for Imbalanced Classification via Deep Reinforcement Learning

Guixuan Wen, Kaigui Wu
2021 Asian Conference on Machine Learning  
This paper proposes a decision tree building method for imbalanced binary classification via deep reinforcement learning.  ...  Next, the reward function is designed based on a range of evaluation metrics of imbalanced classification.  ...  Numerous algorithms have been proposed for imbalanced data classification during the past two decades. They usually can be divided into two groups: the data level and the algorithmic level.  ... 
dblp:conf/acml/WenW21 fatcat:j6ucuk6umngqjeqvgfptiab3uy

Astraea: Self-balancing Federated Learning for Improving Classification Accuracy of Mobile Deep Learning Applications [article]

Moming Duan, Duo Liu, Xianzhang Chen, Yujuan Tan, Jinting Ren, Lei Qiao, Liang Liang
2020 arXiv   pre-print
Compared with FedAvg, the state-of-the-art FL algorithm, Astraea shows +5.59% and +5.89% improvement of top-1 accuracy on the imbalanced EMNIST and imbalanced CINIC-10 datasets, respectively.  ...  divergence (KLD) of their data distribution.  ...  ACKNOWLEDGMENT We would like to thank the anonymous reviewers for their valuable feedback and improvements to this paper. This  ... 
arXiv:1907.01132v2 fatcat:be74enpvpzcebg3eyvi5oa26vm

Efficient Feature Embeddings for Student Classification with Variational Auto-encoders

Severin Klingler, Rafael Wampfler, Tanja Käser, Barbara Solenthaler, Markus H. Gross
2017 Educational Data Mining  
Further, we demonstrate on two independent data sets that our method outperforms previous methods for finding efficient feature embeddings and generalizes better to imbalanced data sets compared to expert  ...  In this paper, we present a semi-supervised classification pipeline that makes effective use of this unlabeled data to significantly improve model quality.  ...  Feature selection proceeds by keeping the K dimensions zi with the largest KL divergence. Semi-supervised classification pipeline.  ... 
dblp:conf/edm/KlinglerWKSG17 fatcat:fhaqbnr6mzcy7gtlnz3cqvxqxu

A ResNet-LSTM Based Credit Scoring Approach for Imbalanced Data

Anqin Zhang, Baicheng Peng, Jingjing Chen, Qingfu Liu, Shibo Jiang, Youmei Zhou
2022 Mobile Information Systems  
The ResNet-LSTM+XGBoost model for classification is better than other traditional algorithms in F1 value, AUC, and KS value, which proves the effectiveness and portability of this model in the field of  ...  Detecting potential defaults or bad debt with limited information has become a huge challenge.  ...  Comparing results with other imbalanced data sample generation algorithms shows that the model training convergence speed and classification effect are improved.  ... 
doi:10.1155/2022/9103437 doaj:ec4bef1c9d424e9cb5c9d612bc67bf99 fatcat:maixroxbabfmvfdbzwiyvl22wy

ECG Arrhythmias Detection Using Auxiliary Classifier Generative Adversarial Network and Residual Network

Pu Wang, Borui Hou, Siyu Shao, Ruqiang Yan
2019 IEEE Access  
class-imbalanced dataset.  ...  We establish classification model based on stacked residual network parallel connected with long short-term memory (LSTM) network.  ...  In this study, ED, PCC, and K-L divergence are chosen as quantitative indicators.  ... 
doi:10.1109/access.2019.2930882 fatcat:mijlqmn6xjbu3pcj5cs2ly5724

Employing One-Class SVM Classifier Ensemble for Imbalanced Data Stream Classification [chapter]

Jakub Klikowski, Michał Woźniak
2020 Lecture Notes in Computer Science  
The classification of imbalanced data streams is gaining more and more interest.  ...  The main idea is to supply the committee with one-class classifiers trained on clustered data for each class separately.  ...  Conclusions We proposed an imbalanced data streams classification algorithm based on the one-class classifier ensemble.  ... 
doi:10.1007/978-3-030-50423-6_9 fatcat:ytf33ign6bcurnmbcek5dn22ym

Classification of Imbalanced Data Using Deep Learning with Adding Noise

Wan-Wei Fan, Ching-Hung Lee, Binghua Cao
2021 Journal of Sensors  
Our approach performs well with smaller network structure compared with other deep models. In addition, the performance is improved over 40% in defective accuracy by adding noise approach.  ...  Besides, a hybrid loss function of crossentropy and KL divergence is proposed. The proposed approach can improve the accuracy of minority class in the testing data.  ...  Acknowledgments This work was supported in part by the Ministry of Science and Technology, Taiwan, under contracts MOST 110-2634-F-009-024, 109-2634-F-009-031, and 109-2218-E-005-015.  ... 
doi:10.1155/2021/1735386 fatcat:aioevowddbbmhbsnhuzhvsoebq

Cost-Sensitive Variational Autoencoding Classifier for Imbalanced Data Classification

Fen Liu, Quan Qian
2022 Algorithms  
Existing algorithms for imbalanced data mainly involve improvements at a single level.  ...  The cost-sensitive VAE classifier was compared with some existing classification algorithms, such as SVM, k-nearest neighbour, decision tree, Bayesian classifier, and random forest.  ... 
doi:10.3390/a15050139 dblp:journals/algorithms/LiuQ22 fatcat:6pumvx736bchfempr4e7n4shh4

Assembly Quality Detection Based on Class-Imbalanced Semi-Supervised Learning

Zichen Lu, Jiabin Jiang, Pin Cao, Yongying Yang
2021 Applied Sciences  
Therefore, we propose a new semi-supervised algorithm that achieves high classification accuracy on the class-imbalanced assembly dataset with limited labeled data.  ...  Based on the mean teacher algorithm, the proposed algorithm uses certainty to select reliable teacher predictions for student learning dynamically, and loss functions are modified to improve the model's  ...  This algorithm improves the mean teacher algorithm and makes up for the deficiency of the semi-supervised learning method in the class-imbalanced scenarios.  ... 
doi:10.3390/app112110373 fatcat:dhzewau3r5ffpml5h7vez5gtuu

A Novel Multiple Unsupervised Algorithm for Land Use/Land Cover Classification

T. Vignesh, K. K. Thyagharajan, D. Murugan, M. Sakthivel, S. Pushparaj
2016 Indian Journal of Science and Technology  
The clustered samples are used as a training set in a novel unsupervised Ensemble Minimization Learning algorithm (EML) for classification.  ...  Improvement: In future it could be apply in video data and could be improve the classification accuracy also.  ...  in Equation v K l m for a neighborhood Klein Graph, G K .  ... 
doi:10.17485/ijst/2016/v9i42/99682 fatcat:5dzwjegkrbczbbplqu76pv2vje

Crowdsourcing with Multiple-Source Knowledge Transfer

Guangyang Han, Jinzheng Tu, Guoxian Yu, Jun Wang, Carlotta Domeniconi
2020 Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence  
As such, via appropriate knowledge transfer, for example, an improved annotation can be achieved for the task at a small cost.  ...  Finally, it adopts an EM algorithm to estimatethe workers strengths and consensus.  ...  Acknowledgements This research was sponsored in part by the U.S. National Science Foundation through Grant Nos. IIS-1763452 and CNS-1828181.  ... 
doi:10.24963/ijcai.2020/398 dblp:conf/ijcai/ShiTZWL20 fatcat:au3id5f7tnbuzi7fd6w5rfwklq

Convolutional Fine-Grained Classification with Self-Supervised Target Relation Regularization [article]

Kangjun Liu, Ke Chen, Kui Jia
2022 arXiv   pre-print
popular fine-grained object benchmarks and superior robustness against sparse and imbalanced data.  ...  Such target coding schemes are less flexible to model inter-class correlation and are sensitive to sparse and imbalanced data distribution as well.  ...  . . . , K and t = 1, 2, . . . , T, the target relation graph matrix G t ∈ R K×K can thus be constructed as the following: G t kl = s(C t−1 k , C t−1 l ) = e Ĉt−1 k • Ĉt−1 l /τ , (6) where G t kl denotes  ... 
arXiv:2208.01997v1 fatcat:bqmo7bziwnfxlk5nvalkitjida
« Previous Showing results 1 — 15 out of 2,866 results