Undersampling Method Research Articles

Undersampling of approaching the classification boundary for imbalance problem

SummaryUsing imbalanced data in classification affect the accuracy. If the classification is based on imbalanced data directly, the results will have large deviations. A common approach to dealing with imbalanced data is to re‐structure the raw dataset via undersampling method. The undersampling method usually uses random or clustering approaches to trimming the majority class in the dataset, since some data in the majority class makes not contribute to classification model. In this paper a revised undersampling approach is proposed. First, we perform space compression in the vertical direction of the separating hyperplane. Then, a weighted random sampling hybrid ensemble learning method is carried out to make the sampled objects spread more widely near the separating hyperplane. Experiments with 7 under‐sampling methods on 21 imbalanced datasets show that our method has achieved good results.

Small molecule drug and biotech drug interaction prediction based on multi-modal representation learning

BackgroundDrug–drug interactions (DDIs) occur when two or more drugs are taken simultaneously or successively. Early detection of adverse drug interactions can be essential in preventing medical errors and reducing healthcare costs. Many computational methods already predict interactions between small molecule drugs (SMDs). As the number of biotechnology drugs (BioDs) increases, so makes the threat of interactions between SMDs and BioDs. However, few computational methods are available to predict their interactions.ResultsConsidering the structural specificity and relational complexity of SMDs and BioDs, a novel multi-modal representation learning method called Multi-SBI is proposed to predict their interactions. First, multi-modal features are used to adequately represent the heterogeneous structure and complex relationships of SMDs and BioDs. Second, an undersampling method based on Positive-unlabeled learning (PU-sampling) is introduced to obtain negative samples with high confidence from the unlabeled data set. Finally, both learned representations of SMD and BioD are fed into DNN classifiers to predict their interaction events. In addition, we also conduct a retrospective analysis.ConclusionsOur proposed multi-modal representation learning method can extract drug features more comprehensively in heterogeneous drugs. In addition, PU-sampling can effectively reduce the noise in the sampling procedure. Our proposed method significantly outperforms other state-of-the-art drug interaction prediction methods. In a retrospective analysis of DrugBank 5.1.0, 14 out of the 20 predictions with the highest confidence were validated in the latest version of DrugBank 5.1.8, demonstrating that Multi-SBI is a valuable tool for predicting new drug interactions through effectively extracting and learning heterogeneous drug features.

Diagnosis of heart failure from imbalance datasets using multi-level classification

Background:The incidence of heart failure is continuing to rise, and the mortality rate is high. Chest X-ray (CXR) has irreplaceable advantages in diagnosing heart failure, such as fast, low risk, and cheap. However, excessive CXR images place a huge burden on physicians and create data imbalance problems. Traditional methods, such as random under-sampling, are used to solve the problems. However, the under-sampling method can destroy the integrity of the data distribution. So, it is necessary to have a method that can address data imbalance problems and assist overburdened healthcare systems. Objective:This study establishes an automatic heart failure diagnosis system based on deep learning from imbalance datasets. Methods:To address the data imbalance problem based on the publicly available CheXpert dataset, this study proposes a method combining under-sampling and instance selection to ensure the integrity of the data distribution. To help physicians better treat heart failure, this study proposes an end-to-end multi-level classification method to diagnose the specific causes of heart failure. Results:On the testing set, our method improves the average accuracy by 3.78% compared to the traditional random under-sampling method, and the accuracy of our end-to-end multi-class classification experiment is 84.44%. Conclusions:The heart failure automated diagnostic system is more efficient and accurate in diagnosing heart failure compared to state-of-the-art methods.

Implementation of Machine Learning Algorithms for Early Detection of Cervical Cancer Based on Behavioral Determinants

Cervical cancer is a disease that affects women and has the highest mortality rate after breast cancer. Early detection of cervical cancer is critical at this time, so cervical cancer patients are decreasing. Many women, especially in Indonesia, are less concerned about the dangers of cervical cancer, even though if detected earlier, this disease will be easier to treat. One alternative for early detection can use machine learning algorithms. The machine learning algorithms used in this study are Naïve Bayes (NB), Logistic Regression (LR), Decision Tree (DT), SVM, and Random Forest. In this study, a random under-sampling method was employed, which had no uses in any prior research. This technique makes the accuracy of the five algorithms even better. The research results show that NB has an accuracy rate of 91.67%, LR has an accuracy rate of 87.5%, DT has an accuracy rate of 81.81%, SVM has an accuracy rate of 75%, and RF has the highest accuracy rate of 94.45%. This research shows that the best model is RF or Random Forest

A Hybrid Evolutionary Under-sampling Method for Handling the Class Imbalance Problem with Overlap in Credit Classification

A Hybrid Evolutionary Under-sampling Method for Handling the Class Imbalance Problem with Overlap in Credit Classification

A Distribution-Preserving Under-Sampling Method for Imbalance Defect Recognition in Castings

Data imbalance is a crucial factor that limits the performance of automatic defect recognition systems in castings. The bias and deterioration of the model are generated by massive normal samples and minor defect samples. Traditional re-sampling methods randomly change the data distribution and ignore the significant intra-class difference among all normal samples. Therefore, this paper proposes a distribution-preserving under-sampling method for imbalance defect-recognition in castings. In detail, our method divides all normal samples into several sub-groups by cluster analysis and reassembles them into some balance datasets, which makes the normal samples in all balance datasets have an identical distribution with the original imbalance dataset. Finally, experiments on our dataset with 3260 images indicate that the proposed method achieves a 0.816 AUC (area under curve) score, which demonstrates significant advantages compared to cost-sensitive learning and re-sampling methods.

Protein-Protein Interaction Sites Prediction Based on an Under-Sampling Strategy and Random Forest Algorithm.

The computational methods of protein-protein interaction sites prediction can effectively avoid the shortcomings of high cost and time in traditional experimental approaches. However, the serious class imbalance between interface and non-interface residues on the protein sequences limits the prediction performance of these methods. This work therefore proposed a new strategy, NearMiss-based under-sampling for unbalancing datasets and Random Forest classification (NM-RF), to predict protein interaction sites. Herein, the residues on protein sequences were represented by the PSSM-derived features, hydropathy index (HI) and relative solvent accessibility (RSA). In order to resolve the class imbalance problem, an under-sampling method based on NearMiss algorithm is adopted to remove some non-interface residues, and then the random forest algorithm is used to perform binary classification on the balanced feature datasets. Experiments show that the accuracy of NM-RF model reaches 87.6% and 84.3% on Dtestset72 and PDBtestset164 respectively, which demonstrate the effectiveness of the proposed NM-RF method in differentiating the interface or non-interface residues.

An Under-Sampling Array Signal Processing Method Based on Improved Hadamard Matrix

The compressive sensing method is an effective way of under-sampling array signal processing, and the measurement matrix is a key technology of compressive sensing processing, which is of decisive significance to improve the performance of under-sampling array signal processing. In this paper, an under-sampling array signal processing method based on an improved Hadamard measurement matrix was proposed. This method improved the construction of the measurement matrix by a compressing zero method capable of reducing the redundant data, enhancing the non-correlation between columns, improving the RIP condition, and improving the performance of the compressive sensing method. The performance was verified by simulation and real data from the lake, and the results showed that the method proposed in this paper has obvious performance advantages compared with the Toeplitz Circular Matrix, the m-sequence-based matrix, and the partial Hadamard matrix under the same conditions.

A novel two-phase clustering-based under-sampling method for imbalanced classification problems

Classification problems with imbalanced data are challenging because traditional classifiers tend to misclassify minority samples. This paper introduces a novel two-phase method in the categories of under-sampling data-level and ensemble-based approaches to tackle these problems. Compared to classic k-means clustering, the main property of our method is that the majority class is partitioned into clusters so that there are no minority samples in the convex-hull of the majority samples of each cluster, and at the same time, the size of each cluster is controlled. Thus, it is expected that by using our method, the general pattern of data in the feature space is kept, and the possibility of changing the data distribution is reduced. Computational results over a variety of imbalanced classification datasets confirm the superiority of our method over the existing methods from different metrics.

CUS-RF-Based Credit Card Fraud Detection with Imbalanced Data

With the continuous expansion of the banks' credit card businesses, credit card fraud has become a serious threat to banking financial institutions. So, the automatic and real-time credit card fraud detection is the meaningful research work. Because machine learning has the characteristics of non-linearity, automation, and intelligence, so that credit card fraud detection can improve the detection efficiency and accuracy. In view of this, this paper proposes a credit card fraud detection model based on heterogeneous ensemble, namely CUS-RF (cluster-based under-sampling boosting and random forest), based on clustering under-sampling and random forest algorithm. CUS-RF-based credit card fraud detection model has the following advantages. Firstly, the CUS-RF model can better overcome the issue of data imbalance. Secondly, based on the idea of heterogeneous ensemble learning, the clustering under-sampling method and random forest model are fused to achieve a better performance for credit card fraud detection. Finally, through the verification of real credit card fraud dataset, the CUS-RF model proposed in this paper has achieved better performance in credit card fraud detection compared with the benchmark model.

Forecasting the yield of wafer by using improved genetic algorithm, high dimensional alternating feature selection and SVM with uneven distribution and high-dimensional data

AbstractWafer yield prediction, as the basis of quality control, is dedicated to predicting quality indices of the wafer manufacturing process. In recent years, data-driven machine learning methods have received a lot of attention due to their accuracy, robustness, and convenience for the prediction of quality indices. However, the existing studies mainly focus on the model level to improve the accuracy of yield prediction does not consider the impact of data characteristics on yield prediction. To tackle the above issues, a novel wafer yield prediction method is proposed, in which the improved genetic algorithm (IGA) is an under-sampling method, which is used to solve the problem of data overlap between finished products and defective products caused by the similarity of manufacturing processes between finished products and defective products in the wafer manufacturing process, and the problem of data imbalance caused by too few defective samples, that is, the problem of uneven distribution of data. In addition, the high-dimensional alternating feature selection method (HAFS) is used to select key influencing processes, that is, key parameters to avoid overfitting in the prediction model caused by many input parameters. Finally, SVM is used to predict the yield. Furthermore, experiments are conducted on a public wafer yield prediction dataset collected from an actual wafer manufacturing system. IGA-HAFS-SVM achieves state-of-art results on this dataset, which confirms the effectiveness of IGA-HAFS-SVM. Additionally, on this dataset, the proposed method improves the AUC score, G-Mean and F1-score by 21.6%, 34.6% and 0.6% respectively compared with the conventional method. Moreover, the experimental results prove the influence of data characteristics on wafer yield prediction.

Network Intrusion Detection with 1D Convolutional Neural Networks

Computer network assets expose to various cyber threats in today’s digital era. Network Anomaly Detection Systems (NADS) play a vital role in protecting digital assets in the purview of network security. Intrusion detection systems data are imbalanced and high dimensioned, affecting models’ performance in classifying malicious traffic. This paper uses a denoising autoencoder (DAE) for feature selection to reduce data dimension. To balance the data, the authors use a combined approach of oversampling technique, adaptive synthetic (ADASYN) and a cluster-based under-sampling method using a clustering algorithm, Kmeans. Then, a one-dimensional convolutional neural network (1D-CNN) is used to perform classification. The performance of the proposed model is evaluated on UNSW-NB15 and NSL-KDD datasets. The experimental results show that the model produces a detection rate of 98.79% and 97.23% on UNSW-NB15 for binary classification and multiclass classification, respectively. In the evaluation using NSL-KDD, the model yields a detection rate of 98.52% for binary type classification and 98.16% for multiclass type classification.

Improved hybrid resampling and ensemble model for imbalance learning and credit evaluation

A clustering-based undersampling (CUS) and distance-based near-miss method are widely used in current imbalanced learning algorithms, but this method has certain drawbacks. In particular, the CUS does not consider the influence of the distance factor on the majority of instances, and the near-miss method omits the inter-class(es) within the majority of samples. To overcome these drawbacks, this study proposes an undersampling method combining distance measurement and majority class clustering. Resampling methods are used to develop an ensemble-based imbalanced-learning algorithm called the clustering and distance-based imbalance learning model (CDEILM). This algorithm combines distance-based undersampling, feature selection, and ensemble learning. In addition, a cluster size-based resampling (CSBR) method is proposed for preserving the original distribution of the majority class, and a hybrid imbalanced learning framework is constructed by fusing various types of resampling methods. The combination of CDEILM and CSBR can be considered as a specific case of this hybrid framework. The experimental results show that the CDEILM and CSBR methods can achieve better performance than the benchmark methods, and that the hybrid model provides the best results under most circumstances. Therefore, the proposed model can be used as an alternative imbalanced learning method under specific circumstances, e.g., for providing a solution to credit evaluation problems in financial applications.

A GAN-based hybrid sampling method for imbalanced customer classification

Class imbalance is a critical issue in customer classification, for which a plethora of techniques have been proposed in the current body of literature. In particular, generative adversarial network (GAN)-based oversampling can capture the true data distribution of minority class samples and generate new samples, and this approach has demonstrated an outstanding ability to address class imbalance. However, GAN-based oversampling suffers from the issue of class overlap. As a result, in this work, we propose a novel a novel GAN-based hybrid sampling method. The new approach first uses GAN-based oversampling to generate the initial balanced dataset and then applies a novel adaptive neighborhood-based weighted undersampling method to remove generated instances and original majority class instances. This approach not only produces instances that fit the actual data distribution but also significantly reduces the influence of class overlap. Experimental results on artificial data and real-world customer datasets show that the proposed GAN-based hybrid sampling method has better performance than other benchmark methods with both accuracy-based and profit-based evaluation metrics.

ProtTrans-Glutar: Incorporating Features From Pre-trained Transformer-Based Models for Predicting Glutarylation Sites.

Lysine glutarylation is a post-translational modification (PTM) that plays a regulatory role in various physiological and biological processes. Identifying glutarylated peptides using proteomic techniques is expensive and time-consuming. Therefore, developing computational models and predictors can prove useful for rapid identification of glutarylation. In this study, we propose a model called ProtTrans-Glutar to classify a protein sequence into positive or negative glutarylation site by combining traditional sequence-based features with features derived from a pre-trained transformer-based protein model. The features of the model were constructed by combining several feature sets, namely the distribution feature (from composition/transition/distribution encoding), enhanced amino acid composition (EAAC), and features derived from the ProtT5-XL-UniRef50 model. Combined with random under-sampling and XGBoost classification method, our model obtained recall, specificity, and AUC scores of 0.7864, 0.6286, and 0.7075 respectively on an independent test set. The recall and AUC scores were notably higher than those of the previous glutarylation prediction models using the same dataset. This high recall score suggests that our method has the potential to identify new glutarylation sites and facilitate further research on the glutarylation process.

De-aliased depth-range-extended optical coherence tomography based on dual under-sampling

We demonstrate a dual under-sampling (DUS) method to achieve de-aliased and depth-range-extended optical coherence tomography (OCT) imaging. The spectral under-sampling can significantly reduce the data size but causes well-known aliasing artifacts. A change in the sampling frequency used to acquire the interference spectrum alters the aliasing period within the output window except for the true image; this feature is utilized to distinguish the true image from the aliasing artifacts. We demonstrate that with DUS, the data size is reduced to 37% at an extended depth range of 24 mm, over which the true depth can be precisely measured without ambiguity. This reduction in data size and precise measuring capability would be beneficial for reducing the acquisition time for OCT imaging in various biomedical and industrial applications.

WDP-BNN: Efficient wafer defect pattern classification via binarized neural network

Wafer map defect pattern classification using convolutional neural network (CNN) has gained a lot of attention in recent years but it demands huge computation and memory cost. Therefore, a WDP-BNN framework based on the binarized neural network is proposed to reduce memory requirement by 29.70× and speed up by 1.66×. To overcome the imbalance problem and performance loss due to binarization of network, advanced data augmentation methods including (Chip Reverse, Chip Translate, Chip Combine) along with random under-sampling method have incorporated in the framework. Experimental results on the WM-811K dataset have demonstrated that the WDP-BNN model has outperformed the state-of-the-art works with the highest classification accuracy of 94.83% and the memory reduction of 1.10-25.93×.

Design of Intelligent Decision System via Computer Aided Translation Software Evaluation under the Background of Internet of Things

The identification and classification of professional terms of machine translation are studied in this work, to improve the accuracy and professionalism of computer aided translation (CAT) software. Firstly, the current situation and related fields of machine translation are analyzed to summarize the difficulties and shortcomings in machine translation. Secondly, the concept of term is introduced to conduct targeted research on the imbalance problem of terminology classification and recognition in machine translation. Thirdly, a term recognition model based on integrated recognition method is proposed. Finally, the classification accuracy and recall rate of the model are verified using the method of confusion matrix in experiments. The results demonstrate that in comparison of the recall rate, classification accuracy, and f value in different fields, the classification accuracy of network terms by the hybrid method combining the over-sampling method and under-sampling method is the highest of 77%, that of sports terms is the lowest of 71%, and that of economic terms is 74%. Among the recall rate, accuracy rate and f value, the recall rate is the highest, reaching more than 80%, especially for economic terms of 91%. The combination of over-sampling and under-sampling performs better than the under-sampling with playback and under-sampling without playback in terms of term recognition and classification in different fields. Through the classification results before and after integration, it is obvious that the integration of each base classifier not only effectively improves the classification accuracy of terms, but also greatly improves the recall rate. This term recognition model can help CAT software in improving the recognition accuracy of term translation, which has certain practical effects and provides reference for research in related fields.

Digital payment fraud detection methods in digital ages and Industry 4.0

The advent of the digital economy and Industry 4.0 enables financial organizations to adapt their processes and mitigate the risks and losses associated with the fraud. Machine learning algorithms facilitate effective predictive models for fraud detection for Industry 4.0. This study aims to identify an efficient and stable model for fraud detection platforms to be adapted for Industry 4.0. By leveraging a real credit card transaction dataset, this study proposes and compares five different learning models: logistic regression, decision tree, k-nearest neighbors, random forest, and autoencoder. Results show that random forest and logistic regression outperform the other algorithms. Besides, the undersampling method and feature reduction using principal component analysis could enhance the results of the proposed models. The outcomes of the studies positively ascertain the effectiveness of using features selection and sampling methods for tackling business problems in the new age of digital economy and industrial 4.0 to detect fraudulent activities.

Classification of pulsar signals using ensemble gradient boosting algorithms based on asymmetric under-sampling method

The detection of pulsar signals is a highly intensive task. Numerous artificial intelligence (AI) and machine learning techniques (ML) have been proposed to classify pulsar and non-pulsar signals. While existing machine learning techniques improve classification efficiency, these methods are limited when it comes to dealing with large volumes of astronomical data, the extreme problem of class imbalance , and the polarization of high recall and precision. In this paper, to accurately classify pulsar and non-pulsar signals, extreme Gradient Boosting (XgBoost), and Light Gradient Boosting Machine (LightGBM) algorithms based on an asymmetric undersampling method are proposed. Firstly, the proposed model uses an asymmetric undersampling technique that divides the benchmark datasets into 75 subsets for the LOFAR Tied-Array All-Sky Survey (LOTAAS-1) and 9 subsets for the High Time Resolution Universe Pulsar Survey (HTRU2), eluding the class imbalance problem. Finally, XgBoost and LightGBM algorithms have been adopted for each new data subset, with a majority voting classifier being used to integrate the output of proposed models against each subset. Results of the proposed method were compared to state-of-the-art baseline models, which showed that our models significantly performed better than existing methods and gained almost 2% and 2.5% better F-score and precision results, respectively.