OPTIMAL SAMPLING RATIO FOR IMBALANCED DATA

Deep learning methods are widely used in the field of rolling bearing fault diagnosis and produce good results when faced with datasets with roughly equal numbers of normal and faulty samples. However, real-world data often has a serious imbalance, with the number of fault samples being significantly less than the number of normal samples. This dataset imbalance challenges the performance of traditional deep learning methods. To address this problem, this paper proposes an efficient imbalanced data rolling bearing fault diagnosis method. The method consists of two parts: a deep learning network based on a multi-scale self-attention mechanism and a novel loss function. In terms of the deep learning network, firstly, the one-dimensional vibration signal is converted into a two-dimensional image through the Gramian angular field. This conversion maximises the inherent feature extraction capability of the network. Subsequently, the multi-scale learning capability of the network is enhanced by implementing different expansion rates for the head of the multi-scale self-attention mechanism. This nuanced approach allows the network to capture the underlying information more efficiently. Finally, the inclusion of Ghost bottlenecks and feature pyramid networks (FPNs) helps to optimise network efficiency and improve generalisation performance. A novel loss function is also proposed to make the method more suitable for imbalanced data. During the training process, the classification of samples in each class is analysed using the recall metric of imbalanced classification and the real-time recall is used as a weight to weaken the dominance of the majority class. This weighting ensures the adaptability of the method to imbalanced datasets. The proposed method is evaluated using rolling bearing datasets from Case Western Reserve University, USA, and Southeast University, China. Comparison results with other state-of-the-art deep learning methods show that the proposed method has a robust performance when dealing with imbalanced data.

An Experimental Assessment of Random Forest Classification Performance Improvisation with Sampling and Stage Wise Success Rate Calculation

The negative impact on learning associated with imbalanced proportion of classes has exploded lately with the exponential growth of “cheap” data. Many real-world problems present scarce number of instances in one class whereas in others their cardinality is several factors greater. The current techniques that treat large-scale imbalanced data are focused on obtaining fast, scalable, and parallel sampling techniques following the standard MapReduce procedure. These generate local balanced solutions in each map, which are eventually combined into a final set. Nevertheless, as we will see later, this divide-and-conquer strategy entails several problems, such as small disjuncts, data lack, etc. In this chapter we also review the latest proposals on imbalanced Big Data preprocessing and present a MapReduce framework for imbalanced preprocessing which includes several state-of-the-art sampling techniques.

The complementary log-log link was originally introduced in 1922 to R. A. Fisher, long before the logit and probit links. While the last two links are symmetric, the complementary log-log link is an asymmetrical link without a parameter associated with it. Several asymmetrical links with an extra parameter were proposed in the literature over last few years to deal with imbalanced data in binomial regression (when one of the classes is much smaller than the other); however, these do not necessarily have the cloglog link as a special case, with the exception of the link based on the generalized extreme value distribution. In this paper, we introduce flexible cloglog links for modeling binomial regression models that include an extra parameter associated with the link that explains some unbalancing for binomial outcomes. For all cases, the cloglog is a special case or the reciprocal version loglog link is obtained. A Bayesian Markov chain Monte Carlo inference approach is developed. Simulations study to evaluate the performance of the proposed algorithm is conducted and prior sensitivity analysis for the extra parameter shows that a uniform prior is the most convenient for all models. Additionally, two applications in medical data (age at menarche and pulmonary infection) illustrate the advantages of the proposedmodels.

Bayesian estimation of abundance based on removal sampling under weak assumption of closed population with catchability depending on environmental conditions. Application to tick abundance

Learning from imbalanced data is an important and common problem. Decision trees, supplemented with sampling techniques, have proven to be an effective way to address the imbalanced data problem. Despite their effectiveness, however, sampling methods add complexity and the need for parameter selection. To bypass these difficulties we propose a new decision tree technique called Hellinger Distance Decision Trees (HDDT) which uses Hellinger distance as the splitting criterion. We analytically and empirically demonstrate the strong skew insensitivity of Hellinger distance and its advantages over popular alternatives such as entropy (gain ratio). We apply a comprehensive empirical evaluation framework testing against commonly used sampling and ensemble methods, considering performance across 58 varied datasets. We demonstrate the superiority (using robust tests of statistical significance) of HDDT on imbalanced data, as well as its competitive performance on balanced datasets. We thereby arrive at the particularly practical conclusion that for imbalanced data it is sufficient to use Hellinger trees with bagging (BG) without any sampling methods. We provide all the datasets and software for this paper online ( http://www.nd.edu/~dial/hddt ).

Learning from the imbalanced data samples so as to achieve accurate classification is an important research content in data mining field. It is very difficult for classification algorithm to achieve a higher accuracy because the uneven distribution of data samples makes some categories have few samples. A imbalanced data classification algorithm of support vector machines (KE-SVM) is proposed in this article, this algorithm achieve the initial classification of data samples by training the maximum margin classification SVM model, and then obtaining a new kernel extension function. based on Chi square test and weight coefficient calculation, through training the samples again by the new vector machine with kernel function to improve the classification accuracy. Through the simulation experiments of real data sets of artificial data set, it shows that the proposed method has higher classification accuracy and faster convergence for the uneven distribution data.

Interactive scribble-and-learning-based segmentation is attractive for its good performance and reduced number of user interaction. Scribbles for foreground and background are often imbalanced. With the arrival of new scribbles, the imbalance ratio may change largely. Failing to deal with imbalanced training data and a changing imbalance ratio may lead to a decreased sensitivity and accuracy for segmentation. We propose a generic Dynamically Balanced Online Random Forest (DyBa ORF) to deal with these problems, with a combination of a dynamically balanced online Bagging method and a tree growing and shrinking strategy to update the random forests. We validated DyBa ORF on UCI machine learning data sets and applied it to two different clinical applications: 2D segmentation of the placenta from fetal MRI and adult lungs from radiographic images. Experiments show it outperforms traditional ORF in dealing with imbalanced data with a changing imbalance ratio, while maintaining a comparable accuracy and a higher efficiency compared with its offline counterpart. Our results demonstrate that DyBa ORF is more suitable than existing ORF for learning-based interactive image segmentation.

In smart homes, data generated from real-time sensors for human activity recognition is complex, noisy and imbalanced. It is a significant challenge to create machine learning models that can classify activities which are not as commonly occurring as other activities. Machine learning models designed to classify imbalanced data are biased towards learning the more commonly occurring classes. Such learning bias occurs naturally, since the models better learn classes which contain more records. This paper examines whether fusing real-world imbalanced multi-modal sensor data improves classification results as opposed to using unimodal data; and compares deep learning approaches to dealing with imbalanced multi-modal sensor data when using various resampling methods and deep learning models. Experiments were carried out using a large multi-modal sensor dataset generated from the Sensor Platform for HEalthcare in a Residential Environment (SPHERE). The data comprises 16104 samples, where each sample comprises 5608 features and belongs to one of 20 activities (classes). Experimental results using SPHERE demonstrate the challenges of dealing with imbalanced multi-modal data and highlight the importance of having a suitable number of samples within each class for sufficiently training and testing deep learning models. Furthermore, the results revealed that when fusing the data and using the Synthetic Minority Oversampling Technique (SMOTE) to correct class imbalance, CNN-LSTM achieved the highest classification accuracy of 93.67% followed by CNN, 93.55%, and LSTM, i.e. 92.98%.

Synthetic Minority Oversampling Technique (SMOTE) is a preferable method used to solve the imbalanced data classification issues. However, its efficiency for resolving the issues of minority sample classification still need to be improved. In order to balance its value and shortcome, we designed a perfected algorithm called “C-K-SMOTE’', which is a mixture clustering algorithm of the Canopy and K-means. For the final purpose of obtaining an approximately balanced data, first we use Canopy to achieve the approximate clustering, then use the K-means to obtain the accurate clustering, and after that we apply the SMOTE to increase the number of minority samples. The referential imbalanced data sets used in the article are selected from KEEL (Knowledge Extraction on Evolutionary Learning). By adopting random forest disaggregated model to carry experiments, SMOTE's efficiency of balancing the imbalanced databases is verified.

Currently, there are many problems in imbalanced big data classification based on rough set with virtual reality technology in cloud computing. For example, redundant big data cleaning is not clear, the effect is poor for big data denoising and feature extraction, and the precision of classification is low. In this paper, an imbalanced big data classification is proposed based on Hubness and K nearest neighbor to address such problems. First, the SNM algorithm is used in order to efficient cleaning of redundant big data. Then, wavelet threshold denoising algorithm is used to denoise the big data to improve the denoising effect. Meantime, feature of big data is extracted based on Lyapunov theorem. Moreover, the Hubness and K-nearest neighbor algorithms are used to achieve high precision of imbalanced big data classification. Experiments verify that the proposed method effectively strengthens current cleaning and denoising methods of redundant imbalanced big data, as well as improves accuracy of extraction and classification of big data.

Aiming at the problem of high‐dimensional features and data imbalance in credit risk assessment that affect model prediction results, a credit risk prediction method based on integrated learning is proposed from three levels of features, data and algorithms. First, a hybrid filter and Random Forest feature selection method is used to select features. This method uses the improved Relief algorithm to initially select features, then combines the maximum information coefficient to eliminate redundant features, and uses the Random Forest algorithm to further reduce the feature dimension. Second, on the basis of the Borderline‐SMOTE method, an adaptive idea is introduced to generate a different number of new samples for each minority sample at the boundary, and a new interpolation method is used to make the new sample distribution more reasonable, so as to reduce the sample imbalance. Finally, the Focal Loss is used to improve the loss function of LightGBM, and the sample weight is adjusted through the parameters α and γ in the Focal Loss function so that the model pays more attention to minority samples and difficult‐to‐classify samples, and improves the accuracy of model classification. And use the improved algorithm as the base classifier and then use AdaBoost and random subspace methods to integrate to establish a credit risk prediction model. Through comparative experiments with other methods, the results show that this method effectively improves the G‐mean value and the AUC value and has a better default prediction effect.

Most existing techniques to handle imbalanced data might be invalid in the presence of data missing since they are based on the assumption that the data is complete. To shorten such a gap, a novel synthetic minority oversampling technique (SMOTE), i.e., Non-negative latent factor analysis-incorporated and Switching triple-weight-SMOTE (NSS), is proposed. The main idea of NSS is 4-fold: 1) a Lagrange non-negative matrix factorization (LNMF) method is put forward to impute the missing values with a guaranteed non-negativity according to the original distribution owing to the consideration of the global feature information; 2) by mapping the complete data after imputation into an empirical feature space (EFS), a more separable dataset is achieved, which rigidly maintains the same geometrical structure as the original data while efficiently reducing the redundant features to enhance model generalization and operational efficiency; 3) after fulfilling the fuzzy <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$c$</tex-math> </inline-formula> -means (FCM) clustering, the inter-cluster distance, the capacity of each minority cluster and its sparsity are comprehensively taken into account to develop a triple-weight assignment strategy, which contributes to allocate the number of synthetic samples to each cluster appropriately; 4) a switching oversampling strategy is provided in response to the clusters with different distributions (i.e., either Gaussian or uniform distribution). Moreover, a posterior is used to check the correctness of the synthetic samples. Finally, experiments on a real dataset and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$12$</tex-math> </inline-formula> public datasets show that the proposed NSS outperforms other <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$11$</tex-math> </inline-formula> state-of-art methods. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Note to Practitioners</i> —Data classification is an important computer task, which has been successfully used in many domains, including but not limited to, the medical domain, finance domain, and manufacturing domain. However, there are two big challenges when a classifier is to handle real-world data in the presence of imbalanced data and missing values. More specifically, the model performance is very likely to be degraded due to the missing information and the inclination of classifiers to the majority class. To surmount this problem, a natural idea is to use the LNMF model to obtain the desired recovery. Then, for the complete data after imputation, the empirical-feature-space-based switching triple-weight-SMOTE is applied to synthesize safe and correct data (i.e., lie solidly on the region of minority class) to achieve the balance. Such a working principle generates a novel NSS strategy. The proposed NSS strategy has the following obvious merits: 1) the imputation guarantees the similarity to the original dataset; and 2) new synthetic data are safely generated by taking adequate consideration of the information and distribution of datasets. Thus, the proposed NSS can greatly help improve the classification accuracy of real-world datasets.

This paper reports the scaling behavior of the fuzzy c-means based classifier (FCMC) with many parameters. FCMC is a classifier based on clustering approaches. The classification accuracy on test sets (i.e., the generalization capability) is not necessarily improved by increasing the number of clusters. Especially when the number of training samples is relatively small, not only the classification boundary over-fits the data, but also covariance matrices and cluster centers are computed incorrectly, since the number of samples in each cluster becomes smaller. Hence, the test set accuracy deteriorates. The performance of FCMC with two clusters in each class and the number of training samples less than 1000, was reported in the literature. This paper reports the scaling behavior of FCMC by testing with variously-sized training samples. The number of clusters of FCMC is increased up to eight. The number of clusters used in this paper is not very large but the number of parameters is relatively large. So, the parameters are optimized to training sets. LibSVM is one of the widely known state of the art tools for support vector machines (SVM). The test set accuracy, training time and testing time (i.e., the detection time) of FCMC are compared with LibSVM by varying the size of training sets. FCMC shows a good generalization capability, though the parameters are optimized to training sets. When the number of training samples is increased by 10 times, the training time of FCMC increases by 10 times, but that of LibSVM increases by a factor of 100. The testing time is also much shorter than LibSVM when the size of the training set is large.

Imbalanced image datasets consist of image datasets where there is a significant disparity in the number of samples across different classes. With imbalanced image datasets, learning algorithms often tend to be biased toward the majority class samples. This leads to poor classification of minority class samples as their training is not properly conducted. It becomes more complicated when the number of samples in the minority class is very low. In this paper, a novel hierarchical algorithm is proposed for generating new data using Capsule Generative Adversarial Networks (CapsGAN) to address the class imbalance problem in imbalanced image datasets. Unlike common GAN models, the proposed method incorporates an auxiliary CapsNet to identify high‐value images in both minority and majority classes. This identification is based on the ability to detect complex relationships between low‐level and high‐level features present in capsule networks. Furthermore, the proposed CapsGAN model is conditioned to generate minority class samples based on feature vectors of last capsule layer to achieve a more balanced image dataset. For evaluating the performance of the proposed model, an image dataset called CICS was collected and introduced. Extensive experiments were also conducted using various online image datasets from different domains, with varying numbers of classes and data sizes. The experimental results demonstrated that the proposed model can generate high‐quality samples in cases where the image dataset or the number of minority class samples is relatively small. Furthermore, the proposed model has maintained an accuracy of over 80% in an imbalanced ratio of 1:60.