Rebalance Weights AdaBoost‐SVM Model for Imbalanced Data

Abstract

Classification of imbalanced data is a challenging task that has captured considerable interest in numerous scientific fields by virtue of the great practical value of minority accuracy. Some methods for improving generalization performance have been developed to address this classification situation. Here, we propose a cost-sensitive ensemble learning method using a support vector machine as a base learner of AdaBoost for classifying imbalanced data. Considering that the existing methods are not well studied in terms of how to precisely control the classification accuracy of the minority class, we developed a novel way to rebalance the weights of AdaBoost, and the weights influence the base learner training. This weighting strategy increases the sample weight of the misclassified minority while decreasing the sample weight of the misclassified majority until their distributions are even in each round. Furthermore, we included P-mean as one of the assessment markers and discussed why it is necessary. Experiments were conducted to compare the proposed and comparison 10 models on 18 datasets in terms of six different metrics. Through comprehensive experimental findings, the statistical study is performed to verify the efficacy and usability of the proposed model.

Analyzing Multilingual Conversations During COVID-19: An Imbalanced Class-Ensemble Learning Approach with Reweighted AdaBoost-SVM for Code-Switched Text Classification

“…The resulting integrated model outperformed traditional classification models, particularly in enhancing the identification of financially distressed samples. Parallel to these developments, Piao et al [45] addressed the challenge of class imbalance in predicting corporate credit risk within supply chain contexts. They adopted a hybrid model that fuses SVM with AdaBoost ensemble models, integrating an artificial imbalance rate model alongside various methods for selecting features.…”

“…These include cost-sensitive AdaBoost adaptations [42], [50] and AdaC2, which have demonstrated effective and stable results [24], [25]. Additionally, heterogeneous SVMs integrated with AdaBoost also yield promising outcomes [45]. An exemplary implementation is the SVM cost-sensitive ensemble based on adaptive cost weights [38], a method that dynamically accounts for the varying impact of positive cases on the SVM classifier during the boosting process, as informed by previously obtained classifiers.…”

“…The penalty factor, symbolized by C, is an established constant in the optimization equation. To derive a classifier, one must initially select a C value, appraise the classifier's effectiveness, and iteratively refine C to attain optimal outcomes [45]. This iterative refinement stands apart from the primary optimization activity.…”

Identification of Pre-Emptive Biosecurity Zone Areas for Highly Pathogenic Avian Influenza Based on Machine Learning-Driven Risk Analysis

“…To address this challenge, we employ a strategy involving the adjustment of class-specific weights [24]. Through iterative testing, we optimize these weights to balance the training process [25]. SVM training is subsequently performed using a Gaussian kernel to improve the model's ability to generalize from the training data to unseen instances.…”

Ensemble convolutional neural network modeling based on a cost-sensitive approach

“…The weight adjustment did not consider the significance of the minority class, meaning that if the minority class is misclassified, more weight should be given to balance the numerical advantage of the majority class. This idea has been applied and proven effective in AdaBoost-SVM [27]. This study successfully combined the weight considerations of the costsensitive AdaBoost framework with the loss function of a CNN-based classifier, harmonizing the balance between sample data weights and classifier accuracy.…”