Artificial Datasets Research Articles

A verified training support vector machine in bearing fault diagnosis

Abstract The combination of support vector machine (SVM) and deep learning is widely used in bearing fault diagnosis. The SVM-based bearing fault diagnosis method usually uses features extracted from bearing vibration signals as input. However, environmental noise in industrial applications can affect the feature representation, potentially leading to failures in SVM-based fault diagnosis methods. Adversarial robustness verification methods can evaluate the robustness of models to small perturbations in feature representations. Certified defense methods achieve robustness enhancement by embedding adversarial robustness verification into the learning process of the model. Thus, we propose a certified defense method named verified training SVM (VT-SVM) based on Lagrange duality. The proposed method incorporates adversarial robustness verification into the SVM learning framework, thereby enhancing the robustness of SVM in noisy environments. First, the certified defense optimization problem of SVM is constructed by combining the original optimization problem of SVM with the adversarial robustness verification problem. Then, the certified defense optimization problem is converted into an equivalent convex quadratic programming problem by introducing slack variables. Next, the Lagrange duality is used to transform the convex quadratic programming problem into a dual problem. Finally, the solution to the original optimization problem can be obtained by solving the dual problem. We validate the effectiveness of VT-SVM on the artificial dataset, the rolling bearing dataset from Case Western Reserve University, and the VBL-VA001 dataset from Sepuluh Nopember Institute of Technology. Experimental results demonstrate that our method both ensures high predictive performance and improves the resistance of the model to small noise.

Simultaneous predictive bands for functional time series using minimum entropy sets

Functional Time Series (FTS) are sequences of dependent random elements taking values on some functional space. Most of the research on this domain focuses on producing a predictor able to forecast the next function, having observed a part of the sequence. For this, the Autoregressive Hilbertian process is a suitable framework. Here, we address the problem of constructing simultaneous predictive confidence bands for a stationary FTS. The method is based on an entropy measure for stochastic processes. To construct predictive bands, we use a Reproducing Kernel Hilbert Space (RKHS) to represent the functions and a functional bootstrap procedure that allows us to estimate the prediction law and a Reproducing Kernel Hilbert Spaces (RKHS) to represent the functions, considering then the basis associated to the reproducing kernel. We then classify the points on the projected space according to those that belong to the minimum entropy set (MES) and those that do not. We map the minimum entropy set back to the functional space and construct a band using the regularity property of the RKHS. The proposed methodology is illustrated through artificial and real data sets.

BERT-Inspired Progressive Stacking to Enhance Spelling Correction in Bengali Text

Common spelling checks in the current digital era have trouble reading languages such as Bengali, which employ English letters differently. In response, we have created a better Bidirectional Encoder Representations from Transformers (BERT)–based spell checker that makes use of a convolutional neural network (CNN) sub-model (Semantic Network). Our novelty, which we term progressive stacking , concentrates on improving BERT model training while expediting the corrective process. We discovered that, when comparing shallow and deep versions, deeper models could require less training time. There is potential for improving spelling corrections with this technique. We categorized and utilized as a test set a 6,300-word dataset that Nayadiganta Mohiuddin supplied, some of which had spelling errors. The most popular terms were the same as those found in the Prothom-Alo artificial error dataset.

Research on concept drift algorithm based on evolutionary computation

Concept drift (CD) in data streams can significantly impact the performance and stability of data stream classification algorithms, diminishing the generalization capabilities of integrated learning models. This paper addresses CD issues in dichotomous data streams by introducing a novel modeling approach that leverages evolutionary computation techniques. The method entails grouping base learners based on their performance within a sliding window and then evolving the base learning periods using evolutionary techniques. Furthermore, the concept of “gene flow” is introduced to enhance diversity among base learners and improve CD prediction performance. Experimental results on real and artificial datasets demonstrate the superior comprehensive performance of the proposed method. Specifically, the BCDECA algorithm outperforms other similar methods, excelling in accuracy, diversity, convergence rate, and robustness on a range of datasets.

Projection generalized correntropy twin support vector regression

A projection generalized maximum correntropy twin support vector regression algorithm is proposed. The generalized correntropy function is added into the loss function of adaptive filtering, maximizing which can suppress the interference of noise or outliers.Considering the fact that single-shift projection twin support vector regression cannot observe local information of samples, a complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) combined with wavelet soft threshold denoising is used to assign weights to samples. The CEEMDAN is used to decompose the original data, calculate the Pearson correlation coefficient between the mode functions and the original data. The mode with low correlation is filtered by wavelet based algorithm with soft-threshold to get the reconstructed samples after noise reduction. Smaller weights will be assigned to reconstructed samples with significant differences from the original data, while larger weights will be assigned to reconstructed samples with smaller differences. Similarly, the empirical risk term in the cost function is also assigned calculated weights to improve the robustness. Due to the use of empirical mode decomposition, the proposed method is particularly suitable for processing non-stationary data. Experimental results on artificial and UCI datasets verified the effectiveness of the algorithm.

Integrating prior field knowledge as key documents with main path analysis utilizing key-node path search

The integration of prior field knowledge in analytical or modeling processes is generally considered favorable across various disciplines, yet its utilization in Main Path Analysis (MPA) has been limited to gathering documents or validating the obtained main paths (MPs). This study envisions that prior knowledge about a field can be embodied in certain key documents that are considered seminal or crucial to the field's development. A so-called key-node path search is then employed to produce MPs that capture a distinct knowledge flow centering around these key documents. This study further proposes a unified approach that automatically and simultaneously produces the key-document MPs alongside the traditional MPs. Through this unified approach, the focused knowledge flow through the key documents and the field's overall knowledge flow, as revealed by the traditional MPs, can be concurrently observed to see how they interact, thereby providing additional insights into the field's development. Not only may the key-document MPs capture a meaningful development trajectory, but their complement to the traditional MPs can also hint at their respective representativeness. To establish this unified approach, this study formally demonstrates how the traditional MPs can be produced with key-node path searches, enabling their simultaneous creation alongside the key-document MPs. A case study is conducted based on patents in the field of Evolutionary Computation from an official artificial intelligence patent dataset to demonstrate the application of this unified approach.

Parameter extension of q-divergence-based fuzzy c-means clustering

AbstractIn this study, we propose a fuzzy clustering algorithm for vectorial data, which is constructed by extending the fuzzification parameters in the q-divergence-based fuzzy c-means algorithm (QFCM). The proposed algorithm, referred to as extended QFCM (eQFCM), is an extension of both QFCM and the penalized fuzzy c-means algorithm proposed by Yang, referred to as Y-type FCM (YFCM). eQFCM extends both the two-parameter QFCM and YFCM algorithms to a four-parameter model. Through numerical experiments using an artificial dataset, we substantiate the theoretical discussion, and the effects of fuzzification parameter to clustering results are observed. Furthermore, the results of some numerical experiments using real datasets are presented to demonstrate that the proposed algorithm outperformed both QFCM and YFCM algorithms in terms of clustering accuracy.

A distribution-free adaptive CUSUM-sign chart for monitoring shifts in the location of unknown industrial process

In most of the manufacturing industry, the distributional assumption of the online working processes is customarily unknown or hard to meet. In such situations, parametric charting mechanisms generally designed under normality assumptions of the model yield more false alarms and invalid out-of-control (OOC) comparisons. The nonparametric (distribution-free) charts are a better choice for practitioners in such cases as their in-control (IC) run length (RL) profiles remain the same. The study intends to develop a new distribution-free adaptive CUSUM sign (NPACUSUM-SN; named hereafter) chart for monitoring the process location. The proposed NPACUSUM-SN chart estimates the unknown process mean shift using an unbiased function and updates adaptively the reference parameter of the CUSUM statistic. The IC robustness of the conventional adaptive CUSUM and the proposed NPACUSUM-SN charts under the symmetric, skewed, and contaminated normal (CN) distributions have been computed using Monte Carlo simulations. The OOC RL profiles of the proposed study have been assessed for the initial state and the shift delay (steady state) in the processes. The proposed NPACUSUM-SN chart shows more resistance against non-normality and effective behaviour as compared to its conventional competitor. The proposed NPACUSUM-SN chart provides uniformly efficient RL characteristics as compared to its counterparts. Implementation of the proposal is made by using a manufacturing industry dataset along with an artificial dataset.

Revisiting Fold-Change Calculation: Preference for Median or Geometric Mean over Arithmetic Mean-Based Methods.

Background: Fold change is a common metric in biomedical research for quantifying group differences in omics variables. However, inconsistent calculation methods and inadequate reporting lead to discrepancies in results. This study evaluated various fold-change calculation methods aiming at a recommendation of a preferred approach. Methods: The primary distinction in fold-change calculations lies in defining group expected values for log ratio computation. To challenge method interchangeability in a "stress test" scenario, we generated diverse artificial data sets with varying distributions (identity, uniform, normal, log-normal, and a mixture of these) and compared calculated fold-changes to known values. Additionally, we analyzed a multi-omics biomedical data set to estimate to what extent the findings apply to real-world data. Results: Using arithmetic means as expected values for treatment and reference groups yielded inaccurate fold-change values more frequently than other methods, particularly when subgroup distributions and/or standard deviations differed significantly. Conclusions: The arithmetic mean method, often perceived as standard or picked without considering alternatives, is inferior to other definitions of the group expected value. Methods using median, geometric mean, or paired fold-change combinations are more robust against violations of equal variances or dissimilar group distributions. Adhering to methods less sensitive to data distribution without trade-offs and accurately reporting calculation methods in scientific reports is a reasonable practice to ensure correct interpretation and reproducibility.

One or two things we know about concept drift-a survey on monitoring in evolving environments. Part B: locating and explaining concept drift.

In an increasing number of industrial and technical processes, machine learning-based systems are being entrusted with supervision tasks. While they have been successfully utilized in many application areas, they frequently are not able to generalize to changes in the observed data, which environmental changes or degrading sensors might cause. These changes, commonly referred to as concept drift can trigger malfunctions in the used solutions which are safety-critical in many cases. Thus, detecting and analyzing concept drift is a crucial step when building reliable and robust machine learning-driven solutions. In this work, we consider the setting of unsupervised data streams which is highly relevant for different monitoring and anomaly detection scenarios. In particular, we focus on the tasks of localizing and explaining concept drift which are crucial to enable human operators to take appropriate action. Next to providing precise mathematical definitions of the problem of concept drift localization, we survey the body of literature on this topic. By performing standardized experiments on parametric artificial datasets we provide a direct comparison of different strategies. Thereby, we can systematically analyze the properties of different schemes and suggest first guidelines for practical applications. Finally, we explore the emerging topic of explaining concept drift.

Dynamic selection for reconstructing instance-dependent noisy labels

As an inevitable issue in annotating large-scale datasets, instance-dependent label noise (IDN) can cause serious overfitting in neural networks. To combat IDN, label reconstruction methods have been developed with noise transition matrices or DNNs to simulate the transition from clean labels to noisy labels. Nevertheless, the absence of correct supervisions will lead to learning wrong noise transitions. This motivates us to select samples with clean labels to fetch the correct supervisions. However, the difficulty in obtaining prior knowledge of the noise rate prohibits the use of existing sample selection methods. To this end, we propose a dynamic sample selection method, namely Identity Mapping (IdMap), to overcome this limitation. Inspired by the feature-dependent characteristic of IDN, we first introduce the extracted instance features and pseudo-ground-truth labels to reconstruct noisy labels. A partial identity mapping between two labels is then established and samples with consistent identity mapping output are selected as clean data to update the classifier. Extensive experiments on both artificial and real-world noisy datasets demonstrate the superiority of IdMap compared with other state-of-the-art methods.

Power to the researchers: Calculating power after estimation

AbstractThis study demonstrates a simple and reliable method for calculating ex post power. We first conduct a series of Monte Carlo experiments to assess its performance. The experiments are designed to produce artificial datasets that resemble actual data from 23 studies funded by the International Initiative for Impact Evaluation (3ie). After determining that the method performs adequately, we then apply it to the 23 studies and compare their ex post power with the ex‐ante power claimed on their funding applications. We find the average ex post power of the 3ie studies is close to 80%. However, there are more estimates of low power than would be expected if all studies had 80% true power. Most of the differences between ex post and ex ante power can be explained by differences between planned and actual total observations, number of clusters, and the degree of intracluster correlation. This demonstrates how ex post power can be used by funders to evaluate previously funded research and identify areas for improved power estimation in future research. We further show how ex post power can aid in the interpretation of both insignificant and significant estimates.

Permutation-invariant linear classifiers

Invariant concept classes form the backbone of classification algorithms immune to specific data transformations, ensuring consistent predictions regardless of these alterations. However, this robustness can come at the cost of limited access to the original sample information, potentially impacting generalization performance. This study introduces an addition to these classes—the permutation-invariant linear classifiers. Distinguished by their structural characteristics, permutation-invariant linear classifiers are unaffected by permutations on feature vectors, a property not guaranteed by other non-constant linear classifiers. The study characterizes this new concept class, highlighting its constant capacity, independent of input dimensionality. In practical assessments using linear support vector machines, the permutation-invariant classifiers exhibit superior performance in permutation experiments on artificial datasets and real mutation profiles. Interestingly, they outperform general linear classifiers not only in permutation experiments but also in permutation-free settings, surpassing unconstrained counterparts. Additionally, findings from real mutation profiles support the significance of tumor mutational burden as a biomarker.

A dyeing clustering algorithm based on ant colony path-finding mechanism

Clustering has a wide range of applications in various industries. In different research ideas, bionic algorithms have heuristic help for clustering research. However, most of the clustering algorithms based on bionics often have problems of insufficient information extraction, too many required parameters, and limited application scope. Therefore, these algorithms have lower accuracy than traditional clustering algorithms. To solve these problems, this paper proposes a new ant colony clustering algorithm, which is called Dyeing Clustering Algorithm based on Ant Colony Path-finding. This algorithm combines the idea of dyeing with pheromones for the first time and realizes clustering by using ants to crawl and dye each point, bringing new inspiration to the artificial intelligence research. This can better capture data features, improve fault tolerance and achieve a better clustering effect. In addition, the algorithm is constructed as a dynamic system to deeply mine data information with the strategy of swarm intelligence, including the ant colony generation model, path-finding model, path-finding termination model, and dyeing clustering model. Besides, to evaluate the algorithm, we selected 2 simulation data sets, 5 artificial data sets, 5 real-world data sets, 13 comparison algorithms, and 8 evaluation indexes. From the comprehensive multi-round experiments, seven of the eight evaluation indexes of the Dyeing Clustering Algorithm based on Ant Colony Path-finding algorithm are more advantageous than other comparison algorithms, which demonstrates its enormous potential in industrial applications.

Large-Scale Indoor Camera Positioning Using Fiducial Markers.

Estimating the pose of a large set of fixed indoor cameras is a requirement for certain applications in augmented reality, autonomous navigation, video surveillance, and logistics. However, accurately mapping the positions of these cameras remains an unsolved problem. While providing partial solutions, existing alternatives are limited by their dependence on distinct environmental features, the requirement for large overlapping camera views, and specific conditions. This paper introduces a novel approach to estimating the pose of a large set of cameras using a small subset of fiducial markers printed on regular pieces of paper. By placing the markers in areas visible to multiple cameras, we can obtain an initial estimation of the pair-wise spatial relationship between them. The markers can be moved throughout the environment to obtain the relationship between all cameras, thus creating a graph connecting all cameras. In the final step, our method performs a full optimization, minimizing the reprojection errors of the observed markers and enforcing physical constraints, such as camera and marker coplanarity and control points. We validated our approach using novel artificial and real datasets with varying levels of complexity. Our experiments demonstrated superior performance over existing state-of-the-art techniques and increased effectiveness in real-world applications. Accompanying this paper, we provide the research community with access to our code, tutorials, and an application framework to support the deployment of our methodology.

Fractional Adaptive Resonance Theory (FRA-ART): An Extension for a Stream Clustering Method with Enhanced Data Representation

Clustering data streams has become a hot topic and has been extensively applied to many real-world applications. Compared with traditional clustering, data stream clustering is more challenging. Adaptive Resonance Theory (ART) is a powerful (online) clustering method, it can automatically adjust to learn both abstract and concrete information, and can respond to arbitrarily large non-stationary databases while having fewer parameters, low computational complexity, and less sensitivity to noise, but its limited feature representation hinders its application to complex data streams. In this paper, considering its advantages and disadvantages, we present its flexible extension for stream clustering, called fractional adaptive resonance theory (FRA-ART). FRA-ART enhances data representation by fractionally exponentiating input features using self-interactive basis functions (SIBFs) and incorporating feature interaction through cross-interactive basis functions (CIBFs) at the cost only of introducing an additionally adjustable fractional order. Both SIBFs and CIBFs can be precomputed using existing algorithms, making FRA-ART easily adaptable to any ART variant. Finally, comparative experiments on five data stream datasets, including artificial and real-world datasets, demonstrate FRA-ART’s superior robustness and comparable or improved performance in terms of accuracy, normalized mutual information, rand index, and cluster stability compared to ART and the state-of-the-art G-Stream algorithm.

BT24 Pseudonymization for artificial intelligence skin lesion datasets: a real-world feasibility study

BT24 Pseudonymization for artificial intelligence skin lesion datasets: a real-world feasibility study

A data-efficient and easy-to-use lip language interface based on wearable motion capture and speech movement reconstruction.

Lip language recognition urgently needs wearable and easy-to-use interfaces for interference-free and high-fidelity lip-reading acquisition and to develop accompanying data-efficient decoder-modeling methods. Existing solutions suffer from unreliable lip reading, are data hungry, and exhibit poor generalization. Here, we propose a wearable lip language decoding technology that enables interference-free and high-fidelity acquisition of lip movements and data-efficient recognition of fluent lip language based on wearable motion capture and continuous lip speech movement reconstruction. The method allows us to artificially generate any wanted continuous speech datasets from a very limited corpus of word samples from users. By using these artificial datasets to train the decoder, we achieve an average accuracy of 92.0% across individuals (n = 7) for actual continuous and fluent lip speech recognition for 93 English sentences, even observing no training burn on users because all training datasets are artificially generated. Our method greatly minimizes users' training/learning load and presents a data-efficient and easy-to-use paradigm for lip language recognition.

Copula-based regression estimation in the presence of outliers

Ordinary least square estimators for linear regression models are highly influenced by outliers. Copula is a powerful tool for studying the dependence between response and predictor variables and may overcome several limitations associated with the classical linear regression models. In this study, we investigate the efficacy of the copula-based regression models over classical linear regression in the presence of outliers in the x, y, and x-y directions. We also examine the performance of several robust regression estimation techniques when variables are related via elliptical and Archimedean family of copulas and outliers are present in the data. Finally, real and artificial data sets are analyzed to compare the performance of various robust regression techniques.

Telemetry Monitoring System with Features Explaining Anomalies Based on Mahalanobis Distance

Because satellites cannot be repaired once launched, operators must detect anomalies early and prevent failures before they occur. Thus, satellite telemetry monitoring systems need to alert operators of anomalies and provide them with useful information to deal with these anomalies. However, traditional knowledge-based monitoring systems have the challenges of difficulty in building comprehensive models and a high dependency on experts. In recent years, data-driven approaches have been actively studied with the development of machine learning algorithms. These approaches solve the challenges of knowledge-based methods; however, they are often less capable of explaining anomalies than knowledge-based methods. In this study, we propose the new telemetry monitoring system with feature engineering to explain anomalies. The proposed method realizes identifiability of anomaly types and unusual telemetry by designing features based on moving averages, telemetry periods, waveform differences, and the Mahalanobis distance. We applied the proposed features to artificial and practical abnormal datasets and evaluated their usefulness. The results showed that the proposed method is capable of identifying trend, periodic, and waveform anomalies, specifying the telemetry in which the anomaly occurred and providing the information to operators.