Multiclass SVM with ramp loss for imbalanced data classification

Phoungphol, Piyaphol; Zhang, Yanqing; Zhao, Yichuan; Srichandan, Bismita

doi:10.1109/grc.2012.6468599

Cited by 11 publications

(5 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The classifier trained with imbalanced data set tend to bias towards the categories with more data and cannot achieve the best results. The higher the degree of data imbalance, the lower the classification accuracy of trained classifier will be [49]. However, the problems of utilizing oversampling techniques include generating unreal data and over-fitting.…”

Section: Smote-enn Methodsmentioning

confidence: 99%

Sub-Macroscopic Inclusion Classification in Bearing Steels Based on LFCN and Ultrasonic Testing

Zhang

Wang

Chen

2023

Preprint

View full text Add to dashboard Cite

With the improvement of science and technology, the demand for advanced steel with excellent performance has gradually increased. Therefore, the evaluation of steel internal cleanness is an important indicator for the evaluation of material quality. Sub-macroscopic inclusions, which size from 50um to 400um and cannot be detected under the domestic and international bearing steel testing standard, are bound to affect the quality, stability and service life of bearing steel seriously. Hence, the researches of inclusion control technology has gradually attracted attention in the academia and industrial manufacture field. In this paper, we propose an end-to-end LFCN classification model based on LSTM unit and 1DFCN, and verify the effectiveness on the large-scale sub-macroscopic inclusion signal data set collected by ultrasonic experiments. To the best of our knowledge, this study is the first one in this field that has acquire such large amount of experimental sub-macroscopic signal data and solve the classification task by FCN. Especially, our framework can accurately detect the features of sub-macroscopic inclusions, which meets the urgent need of the metallurgical industry. The accuracy rate of proposed model is 88.97%, which is state-of-the-art experimental result among other strong time series classifiers.

show abstract

Section: Smote-enn Methodsmentioning

confidence: 99%

Sub-Macroscopic Inclusion Classification in Bearing Steels Based on LFCN and Ultrasonic Testing

Zhang

Wang

Chen

2023

Preprint

View full text Add to dashboard Cite

show abstract

“…by assigning a higher cost for minority classes, which is similar to modify the loss function [22]). In this section, we propose to apply this idea to SVORIM, in order to compare the results obtained by the proposed over-sampling methods to a cost-sensitive approach.…”

Section: Cost-sensitive Svorim (Cs-svorim)mentioning

confidence: 98%

Graph-Based Approaches for Over-Sampling in the Context of Ordinal Regression

Pérez-Ortiz

Gutiérrez

Hervás-Martı́nez

et al. 2015

IEEE Trans. Knowl. Data Eng.

View full text Add to dashboard Cite

The classification of patterns into naturally ordered labels is referred to as ordinal regression or ordinal classification. Usually, this classification setting is by nature highly imbalanced, because there are classes in the problem that are a priori more probable than others. Although standard over-sampling methods can improve the classification of minority classes in ordinal classification, they tend to introduce severe errors in terms of the ordinal label scale, given that they do not take the ordering into account. A specific ordinal over-sampling method is developed in this paper for the first time in order to improve the performance of machine learning classifiers. The method proposed includes ordinal information by approaching over-sampling from a graph-based perspective. The results presented in this paper show the good synergy of a popular ordinal regression method (a reformulation of support vector machines) with the graph-based proposed algorithms, and the possibility of improving both the classification and the ordering of minority classes. A cost-sensitive version of the ordinal regression method is also introduced and compared with the over-sampling proposals, showing in general lower performance for minority classes.

show abstract

“…These functions, while yielding accurate result, still suffer from being computationally-expensive when solved as an optimization problem due to its non-convex nature. Alternatively, a truncated hinge loss or ramp loss function which is a non-smooth but continuous function has been proved to be accurate and efficient for SVM problem [26][27][28] . Due to its efficiency and simplicity, we decide to use the ramp loss to simulate 0-1 loss function in Eq.…”

Section: Multiclass Classification With Ramp Lossmentioning

confidence: 99%

Robust multiclass classification for learning from imbalanced biomedical data

Phoungphol

Zhang

Zhao

2012

Tinshhua Sci. Technol.

Self Cite

View full text Add to dashboard Cite

Imbalanced data is a common and serious problem in many biomedical classification tasks. It causes a bias on the training of classifiers and results in lower accuracy of minority classes prediction. This problem has attracted a lot of research interests in the past decade. Unfortunately, most research efforts only concentrate on 2-class problems. In this paper, we study a new method of formulating a multiclass Support Vector Machine (SVM) problem for imbalanced biomedical data to improve the classification performance. The proposed method applies cost-sensitive approach and ramp loss function to the Crammer and Singer multiclass SVM formulation. Experimental results on multiple biomedical datasets show that the proposed solution can effectively cure the problem when the datasets are noisy and highly imbalanced.

show abstract

Multiclass SVM with ramp loss for imbalanced data classification

Cited by 11 publications

References 15 publications

Sub-Macroscopic Inclusion Classification in Bearing Steels Based on LFCN and Ultrasonic Testing

Sub-Macroscopic Inclusion Classification in Bearing Steels Based on LFCN and Ultrasonic Testing

Graph-Based Approaches for Over-Sampling in the Context of Ordinal Regression

Robust multiclass classification for learning from imbalanced biomedical data

Contact Info

Product

Resources

About