One-class Classification Research Articles

Making the Most of Nothing: One-Class Classification for Single-Molecule Transport Studies.

Single-molecule experiments offer a unique means to probe molecular properties of individual molecules-yet they rest upon the successful control of background noise and irrelevant signals. In single-molecule transport studies, large amounts of data that probe a wide range of physical and chemical behaviors are often generated. However, due to the stochasticity of these experiments, a substantial fraction of the data may consist of blank traces where no molecular signal is evident. One-class (OC) classification is a machine learning technique to identify a specific class in a data set that potentially consists of a wide variety of classes. Here, we examine the utility of two different types of OC classification models on four diverse data sets from three different laboratories. Two of these data sets were measured at cryogenic temperatures and two at room temperature. By training the models solely on traces from a blank experiment, we demonstrate the efficacy of OC classification as a powerful and reliable method for filtering out blank traces from a molecular experiment in all four data sets. On a labeled 4,4'-bipyridine data set measured at 4.2 K, we achieve an accuracy of 96.9 ± 0.3 and an area under the receiver operating characteristic curve of 99.5 ± 0.3 as validated over a fivefold cross-validation. Given the wide range of physical and chemical properties that can be probed in single-molecule experiments, the successful application of OC classification to filter out blank traces is a major step forward in our ability to understand and manipulate molecular properties.

Toward Interpretable Cell Image Representation and Abnormality Scoring for Cervical Cancer Screening Using Pap Smears.

Screening is critical for prevention and early detection of cervical cancer but it is time-consuming and laborious. Supervised deep convolutional neural networks have been developed to automate pap smear screening and the results are promising. However, the interest in using only normal samples to train deep neural networks has increased owing to the class imbalance problems and high-labeling costs that are both prevalent in healthcare. In this study, we introduce a method to learn explainable deep cervical cell representations for pap smear cytology images based on one-class classification using variational autoencoders. Findings demonstrate that a score can be calculated for cell abnormality without training models with abnormal samples, and we localize abnormality to interpret our results with a novel metric based on absolute difference in cross-entropy in agglomerative clustering. The best model that discriminates squamous cell carcinoma (SCC) from normals gives 0.908±0.003 area under operating characteristic curve (AUC) and one that discriminates high-grade epithelial lesion (HSIL) 0.920±0.002 AUC. Compared to other clustering methods, our method enhances the V-measure and yields higher homogeneity scores, which more effectively isolate different abnormality regions, aiding in the interpretation of our results. Evaluation using an external dataset shows that our model can discriminate abnormality without the need for additional training of deep models.

Advancing fraud detection in New Zealand Mānuka honey: Integrating hyperspectral imaging and GANomaly-based one-class classification

New Zealand Mānuka honey has become a prime target for adulteration due to its high commercial value. Given the diverse possibilities of fraudulent activities, training a supervised model that exhaustively covers all potential fraud scenarios is challenging. This study presents a new method for detecting fraudulent behavior in New Zealand Mānuka honey by combining hyperspectral imaging (HSI) with the GANomaly-based One-Class Classification method. We collected 18 different UMF-graded pure Mānuka honey samples from five New Zealand brands, which were used for training. The model was tested on fraudulent honey, including aged and syrup-adulterated honey, and compared with the traditional One-Class Classification methods. The results demonstrate that the HSI combined with the GANomaly method achieved 100% discrimination for all test samples, outperforming the standard rival techniques. In conclusion, this research developed a versatile model capable of detecting honey fraudulent behavior, showing significant practical implications for honey quality assessment.

Deep temporal semi-supervised one-class classification for GNSS radio frequency interference detection

Deep temporal semi-supervised one-class classification for GNSS radio frequency interference detection

Absolute Value Inequality SVM for the PU Learning Problem

Positive and unlabeled learning (PU learning) is a significant binary classification task in machine learning; it focuses on training accurate classifiers using positive data and unlabeled data. Most of the works in this area are based on a two-step strategy: the first step is to identify reliable negative examples from unlabeled examples, and the second step is to construct the classifiers based on the positive examples and the identified reliable negative examples using supervised learning methods. However, these methods always underutilize the remaining unlabeled data, which limits the performance of PU learning. Furthermore, many methods require the iterative solution of the formulated quadratic programming problems to obtain the final classifier, resulting in a large computational cost. In this paper, we propose a new method called the absolute value inequality support vector machine, which applies the concept of eccentricity to select reliable negative examples from unlabeled data and then constructs a classifier based on the positive examples, the selected negative examples, and the remaining unlabeled data. In addition, we apply a hyperparameter optimization technique to automatically search and select the optimal parameter values in the proposed algorithm. Numerical experimental results on ten real-world datasets demonstrate that our method is better than the other three benchmark algorithms.

Chinese satellite frequency and orbit entity relation extraction method based on dynamic integrated learning

Given the scarcity of Satellite Frequency and Orbit (SFO) resources, it holds paramount importance to establish a comprehensive knowledge graph of SFO field (SFO-KG) and employ knowledge reasoning technology to automatically mine available SFO resources. An essential aspect of constructing SFO-KG is the extraction of Chinese entity relations. Unfortunately, there is currently no publicly available Chinese SFO entity Relation Extraction (RE) dataset. Moreover, publicly available SFO text data contain numerous NA (representing for “No Answer”) relation category sentences that resemble other relation sentences and pose challenges in accurate classification, resulting in low recall and precision for the NA relation category in entity RE. Consequently, this issue adversely affects both the accuracy of constructing the knowledge graph and the efficiency of RE processes. To address these challenges, this paper proposes a method for extracting Chinese SFO text entity relations based on dynamic integrated learning. This method includes the construction of a manually annotated Chinese SFO entity RE dataset and a classifier combining features of SFO resource data. The proposed approach combines integrated learning and pre-training models, specifically utilizing Bidirectional Encoder Representation from Transformers (BERT). In addition, it incorporates one-class classification, attention mechanisms, and dynamic feedback mechanisms to improve the performance of the RE model. Experimental results show that the proposed method outperforms the traditional methods in terms of F1 value when extracting entity relations from both balanced and long-tailed datasets.

Distance-based novelty detection model for identifying individuals at risk of developing Alzheimer's disease.

Novelty detection (ND, also known as one-class classification) is a machine learning technique used to identify patterns that are typical of the majority class and can discriminate deviations as novelties. In the context of Alzheimer's disease (AD), ND could be employed to detect abnormal or atypical behavior that may indicate early signs of cognitive decline or the presence of the disease. To date, few research studies have used ND to discriminate the risk of developing AD and mild cognitive impairment (MCI) from healthy controls (HC). In this work, two distinct cohorts with highly heterogeneous data, derived from the Australian Imaging Biomarkers and Lifestyle (AIBL) Flagship Study of Ageing project and the Fujian Medical University Union Hospital (FMUUH) China, were employed. An innovative framework with built-in easily interpretable ND models constructed solely on HC data was introduced along with proposing a strategy of distance to boundary (DtB) to detect MCI and AD. Subsequently, a web-based graphical user interface (GUI) that incorporates the proposed framework was developed for non-technical stakeholders. Our experimental results indicate that the best overall performance of detecting AD individuals in AIBL and FMUUH datasets was obtained by using the Mixture of Gaussian-based ND algorithm applied to single modality, with an AUC of 0.8757 and 0.9443, a sensitivity of 96.79% and 89.09%, and a specificity of 89.63% and 90.92%, respectively. The GUI offers an interactive platform to aid stakeholders in making diagnoses of MCI and AD, enabling streamlined decision-making processes. More importantly, the proposed DtB strategy could visually and quantitatively identify individuals at risk of developing AD.

Harmonizing Compliance: Coordinating Automated Verification Processes within Cloud-based AI/ML Workflows

The significance of ensuring security and upholding data privacy within cloud-based workflows is widely recognized in research domains. This importance is particularly evident in contexts such as safeguarding patients' private data managed within cloud-deployed workflows, where maintaining confidentiality is paramount, alongside ensuring secure communication among involved stakeholders. In response to these imperatives, our paper presents an architecture and formal model designed to enforce security measures within cloud workflow orchestration. Central to our proposed architecture is the emphasis on continuous monitoring of cloud resources, workflow tasks, and data streams to detect and preempt anomalies in workflow orchestration processes. To accomplish this, we advocate for a multi-modal approach that integrates deep learning, one-class classification, and clustering techniques. In essence, our proposed architecture offers a comprehensive solution for enforcing security within cloud workflow orchestration, harnessing advanced methodologies like deep learning for anomaly detection and prediction. This approach is particularly pertinent in critical sectors such as healthcare, especially during unprecedented events like the COVID-19 pandemic.

Harmonizing Compliance: Coordinating Automated Verification Processes within Cloud-based AI/ML Workflows

The significance of ensuring security and upholding data privacy within cloud-based workflows is widely recognized in research domains. This importance is particularly evident in contexts such as safeguarding patients' private data managed within cloud-deployed workflows, where maintaining confidentiality is paramount, alongside ensuring secure communication among involved stakeholders. In response to these imperatives, our paper presents an architecture and formal model designed to enforce security measures within cloud workflow orchestration. Central to our proposed architecture is the emphasis on continuous monitoring of cloud resources, workflow tasks, and data streams to detect and preempt anomalies in workflow orchestration processes. To accomplish this, we advocate for a multi-modal approach that integrates deep learning, one-class classification, and clustering techniques. In essence, our proposed architecture offers a comprehensive solution for enforcing security within cloud workflow orchestration, harnessing advanced methodologies like deep learning for anomaly detection and prediction. This approach is particularly pertinent in critical sectors such as healthcare, especially during unprecedented events like the COVID-19 pandemic.

A PU‐learning based approach for cross‐site scripting attacking reality detection

AbstractCross‐site scripting (XSS) attack has been one of the most dangerous attacks in cyberspace security. Traditional methods essentially discover XSS attack by detecting malicious payloads in requests, which is unable to distinguish attacking attempts with the attacking reality. The authors collect responses from a web server and train a bagging‐based PU learning model to determine whether the XSS vulnerability is truly triggered. To validate the authors’ proposed framework, experiments are performed on 5 popular web applications with 11 specified CVE recorded vulnerabilities and 32 vulnerable inputs. Results show that the authors’ approach outperforms existing research studies, effectively identifies the attacking reality from attacking attempts, and meanwhile reduces the number of worthless security alarms.

Dense Projection for Anomaly Detection

This work presents a novel method called dense projection for unsupervised anomaly detection (DPAD). The main idea is maximizing the local density of (normal) training data and then determining whether a test data is anomalous or not by evaluating its density. Specifically, DPAD uses a deep neural network to learn locally dense representations of normal data. Since density estimation is computationally expensive, we minimize the local distances of the representations in an iteratively reweighting manner, where the weights are updated adaptively and the parameters are regularized to avoid model collapse (all representations collapse to a single point). Compared with many state-of-the-art methods of anomaly detection, our DPAD does not rely on any assumption about the distribution or spatial structure of the normal data and representations. Moreover, we provide theoretical guarantees for the effectiveness of DPAD. The experiments show that our method DPAD is effective not only in traditional one-class classification problems but also in scenarios with complex normal data composed of multiple classes.

Vibration anomaly detection of wind turbine based on temporal convolutional network and support vector data description

Due to the complex working conditions and harsh environment, wind turbines often encounter abnormalities, resulting in great operation and maintenance difficulties. As nacelle vibration signals reveal the structure's dynamic characteristics and the interaction between components, vibration anomaly detection has a strong potential. However, vibration anomaly detection is challenging due to the complex characteristics and non-stationarity of high dynamic nacelle vibration signals. To solve this problem, this paper proposed a semi-supervised vibration anomaly detection approach for wind turbines, combining deep learning and one-class classification. Firstly, a healthy behavior model (HBM) for predicting wind turbine nacelle vibration based on the temporal convolutional network (TCN) is developed. To use all available information, a Hilbert spectrum fusion technology (HSFT) was proposed to enhance model performance. Then, based on the support vector data description (SVDD) algorithm, we established a one-class classifier of vibration prediction residual and realized the vibration anomaly detection. The proposed approach can be trained on the healthy dataset to provide accurate detection of different abnormal types. The effectiveness of the proposed anomaly detection approach was verified on simulated and actual monitoring datasets.

Analysis Of One-Class and Binary Classification For COVID-19 Image Data

With the recent Coronavirus Disease 2019 (COVID-19) outbreak, there has been a growing interest in the utilization of machine learning models to aid in medical diagnostics. However, during the initial phases of the pandemic, the available datasets were severely imbalanced due to a scarcity of positive cases. These datasets predominantly comprised normal cases, with only a limited number of disease instances. This inherent data imbalance restricted the effectiveness of binary classification, prompting an exploration of one-class classification. This study aims to compare the performance of one-class and binary classification models during the early stages of COVID-19 development. The objective is to provide healthcare professionals with the most effective machine learning-assisted solutions for handling future outbreaks of a similar nature. The experiment incorporates three distinct models: One-Class Support Vector Machine (OCSVM) for one-class classification, alongside Categorical Boosting (CatBoost) and Extreme Gradient Boosting (XGBoost) for binary classification. These models undergo a rigorous tuning process employing grid search, using training data composed of features extracted from compressed chest X-ray images. In summary, the research findings unequivocally demonstrate the superior performance of one-class classification over binary classification in all early stages, achieving an impressive F1 score of 0.46, compared to binary classification's modest score of 0.063. Consequently, it is strongly recommended to employ a one-class classification model to enhance real-life diagnostic processes. However, it is anticipated that as more positive cases are incorporated into the dataset, achieving a more balanced distribution, binary classification may exhibit greater potential. Thus, medical practitioners should reconsider the applicability of binary classification in later stages when the dataset attains improved balance.

Data-Driven Version of Multiway Soft Independent Modeling of Class Analogy (N-Way DD-SIMCA): Theory and Application.

One-class classification (OCC) is discussed in the framework of the measurement and processing of multiway data. Data-driven soft independent modeling of class analogy (DD-SIMCA) is applied in the following formats: (1) multiblock and (2) Tucker 3 N-way SIMCA, which are shown to be useful tools for solving classification tasks. A new decision rule for N-way DD-SIMCA is adopted based on the conventional two-way DD-SIMCA model. Multiblock SIMCA is shown to be useful for variable selection, and Tucker 3 SIMCA to select the optimal model complexity when applying multiway data decomposition and to assess the role of individual samples in the classification model. Both approaches, together with the two-way DD-SIMCA version applied to the unfolded data, are compared regarding the analysis of an experimental data set including genuine and adulterated blueberry extract samples. The latter were employed to produce matrix spectral-time data matrices per sample within a flow injection system, taking advantage of the spectral changes in the sample constituents as a function of the pH of the carrier phase. The need to employ the Tucker 3 model instead of a trilinear decomposition is supported by a discussion on the lack of the trilinearity property of the studied data.

Semi-Supervised Disease Classification Based on Limited Medical Image Data.

Inrecent years, significant progress has been made in the field of learning from positive and unlabeled examples (PU learning), particularly in the context of advancing image and text classification tasks. However, applying PU learning to semi-supervised disease classification remains a formidable challenge, primarily due to the limited availability of labeled medical images. In the realm of medical image-aided diagnosis algorithms, numerous theoretical and practical obstacles persist. The research on PU learning for medical image-assisted diagnosis holds substantial importance, as it aims to reduce the time spent by professional experts in classifying images. Unlike natural images, medical images are typically accompanied by a scarcity of annotated data, while an abundance of unlabeled cases exists. Addressing these challenges, this paper introduces a novel generative model inspired by Hölder divergence, specifically designed for semi-supervised disease classification using positive and unlabeled medical image data. In this paper, we present a comprehensive formulation of the problem and establish its theoretical feasibility through rigorous mathematical analysis. To evaluate the effectiveness of our proposed approach, we conduct extensive experiments on five benchmark datasets commonly used in PU medical learning: BreastMNIST, PneumoniaMNIST, BloodMNIST, OCTMNIST, and AMD. The experimental results clearly demonstrate the superiority of our method over existing approaches based on KL divergence. Notably, our approach achieves state-of-the-art performance on all five disease classification benchmarks. By addressing the limitations imposed by limited labeled data and harnessing the untapped potential of unlabeled medical images, our novel generative model presents a promising direction for enhancing semi-supervised disease classification in the field of medical image analysis.

Pinball-OCSVM for Early-Stage COVID-19 Diagnosis with Limited Posteroanterior Chest X-Ray Images

The conventional way of respiratory coronavirus disease 2019 (COVID-19) diagnosis is reverse transcription polymerase chain reaction (RT-PCR), which is less sensitive during early stages; especially if the patient is asymptomatic, which may further cause more severe pneumonia. In this context, several deep learning models have been proposed to identify pulmonary infections using publicly available chest X-ray (CXR) image datasets for early diagnosis, better treatment and quick cure. In these datasets, the presence of less number of COVID-19 positive samples compared to other classes (normal, pneumonia and Tuberculosis) raises the challenge for unbiased learning of deep learning models. This learning problem can be considered as one-class classification problem where the target class samples are present and other classes are absent or ill-defined. All deep learning models opted class balancing techniques to solve this issue; which however should be avoided in any medical diagnosis process. Moreover, the deep learning models are also data hungry and need massive computation resources. Therefore, for quicker diagnosis, this research proposes a novel pinball loss function based one-class support vector machine (PB-OCSVM), that can work in presence of limited COVID-19 positive CXR samples (target class or class-of-interest (CoI) samples) with objectives to maximize the learning efficiency and to minimize the false predictions. The performance of the proposed model is compared with conventional OCSVM and recent deep learning models, and the experimental results prove that the proposed model outperformed state-of-the-art methods. To validate the robustness of the proposed model, experiments are also performed with noisy CXR images and UCI benchmark datasets.

A One-Class Classifier for the Detection of GAN Manipulated Multi-Spectral Satellite Images

The current image generative models have achieved a remarkably realistic image quality, offering numerous academic and industrial applications. However, to ensure these models are used for benign purposes, it is essential to develop tools that definitively detect whether an image has been synthetically generated. Consequently, several detectors with excellent performance in computer vision applications have been developed. However, these detectors cannot be directly applied as they areto multi-spectral satellite images, necessitating the training of new models. While two-class classifiers generally achieve high detection accuracies, they struggle to generalize to image domains and generative architectures different from those encountered during training. In this paper, we propose a one-class classifier based on Vector Quantized Variational Autoencoder 2 (VQ-VAE 2) features to overcome the limitations of two-class classifiers. We start by highlighting the generalization problem faced by binary classifiers. This was demonstrated by training and testing an EfficientNet-B4 architecture on multiple multi-spectral datasets. We then illustrate that the VQ-VAE 2-based classifier, which was trained exclusively on pristine images, could detect images from different domains and generated by architectures not encountered during training. Finally, we conducted a head-to-head comparison between the two classifiers on the same generated datasets, emphasizing the superior generalization capabilities of the VQ-VAE 2-based detector, wherewe obtained a probability of detection at a 0.05 false alarm rate of 1 for the blue and red channels when using the VQ-VAE 2-based detector, and 0.72 when we used the EfficientNet-B4 classifier.

Robust control chart for nonlinear conditionally heteroscedastic time series based on Huber support vector regression.

This study proposes a control chart that monitors conditionally heteroscedastic time series by integrating the Huber support vector regression (HSVR) and the one-class classification (OCC) method. For this task, we consider the model that incorporates nonlinearity to the generalized autoregressive conditionally heteroscedastic (GARCH) time series, named HSVR-GARCH, to robustly estimate the conditional volatility when the structure of time series is not specified with parameters. Using the squared residuals, we construct the OCC-based control chart that does not require any posterior modifications of residuals unlike previous studies. Monte Carlo simulations reveal that deploying squared residuals from the HSVR-GARCH model to control charts can be immensely beneficial when the underlying model becomes more complicated and contaminated with noises. Moreover, a real data analysis with the Nasdaq composite index and Korea Composite Stock Price Index (KOSPI) datasets further disclose the validity of using the bootstrap method in constructing control charts.

PU-Detector: A PU Learning-based Framework for Real Money Trading Detection in MMORPG

Massive multiplayer online role-playing games (MMORPG) have been becoming one of the most popular and exciting online games. In recent years, a cheating phenomenon called real money trading (RMT) has arisen and damaged the fantasy world in many ways. RMT is the sale of in-game items, currency, or even characters to earn real money, breaking the balance of the game economy ecosystem and damaging the game experience. Therefore, some studies have emerged to address the problem of RMT detection. However, they cannot well handle the label uncertainty problem in practice, where there are only labeled RMT samples (positive samples) and unlabeled samples, which could either be RMT samples or normal transactions (negative samples). Meanwhile, the trading relationship between RMTers is modeled in a simple way, leading to some normal transactions being falsely classified as RMT. In this article, we propose PU-Detector, a novel framework based on PU learning (learning from positive and unlabeled data) for RMT detection, considering the fact that there are only labeled RMT samples and other unlabeled transactions. We first automatically estimate the likelihood of one transaction being RMT by developing an improved PU learning method and proposing an assessment rule. Sequentially, we use the estimated likelihood as edge weight to construct a trading graph to learn trader representation. Then, with the trader representations and basic trading features, we detect RMT samples by the improved PU learning method. PU-Detector is evaluated on a large-scale real world dataset consisting of 33,809,956 transaction logs generated by 43,217 unique players. Compared with other approaches, it achieves the state-of-the-art performance and demonstrates its advantages in detecting underlying RMT samples.

Unsupervised Fault Detection With Deep One-Class Classification and Manifold Distribution Alignment

Fault detection or anomaly detection relies heavily on learning from datasets where only normal samples are available, resulting in the emergence of numerous one-class classification (OCC) methods. However, learning discriminative deep representatives with good generalization from cross-domain positive samples remains challenging. Therefore, this work proposes an end-to-end framework, deep transfer one-class classification (DTOCC) for unsupervised fault detection, which combines adversarial generative OCC and distribution alignment from the perspective of manifold learning. Specifically, pseudo-negative samples are generated outside the positive manifold, facilitating the model to learn discrimination with respect to normal and anomaly. Further, cross-domain positive samples are aligned in log-Euclidean manifold space to enhance representation learning. Then, we provide the specific implementations for fault detection and validate its superiority through case studies on multi-class and run-to-failure datasets, simulating both offline and online scenarios.