Data Streams Research Articles

Avian Influenza: Lessons from Past Outbreaks and an Inventory of Data Sources, Mathematical and AI Models, and Early Warning Systems for Forecasting and Hotspot Detection to Tackle Ongoing Outbreaks.

The ongoing avian influenza (H5N1) outbreak, one of the most widespread and persistent in recent history, has significantly impacted public health and the poultry and dairy cattle industries. This review covers lessons from past outbreaks, risk factors for transmission, molecular epidemiology, clinical features, surveillance strategies, and socioeconomic impacts. Since 1997, H5N1 has infected over 900 individuals globally, with a fatality rate exceeding 50%. Key factors influencing infection rates include demographic, socioeconomic, environmental, and ecological variables. The virus's potential for sustained human-to-human transmission remains a concern. The current outbreak, marked by new viral clades, has complicated containment efforts. This review discusses how to integrate technological advances, such as mathematical modeling and artificial intelligence (AI), to improve forecasting, hotspot detection, and early warning systems. We provide inventories of data sources, covering both conventional and unconventional data streams, as well as those of mathematical and AI models, which can be vital for comprehensive surveillance and outbreak responses. In conclusion, integrating AI, mathematical models, and technological innovations into a One-Health approach is essential for improving surveillance, forecasting, and response strategies to mitigate the impacts of the ongoing avian influenza outbreak. Strengthening international collaboration and biosecurity measures will be pivotal in controlling future outbreaks and protecting both human and animal populations from this evolving global threat.

Intensive Class Imbalance Learning in Drifting Data Streams

Intensive Class Imbalance Learning in Drifting Data Streams

Continual learning with high-order experience replay for dynamic network embedding

Dynamic network embedding (DNE) poses a tough challenge in graph representation learning, especially when confronted with the frequent updates of streaming data. Conventional DNEs primarily resort to parameter updating but perform inadequately on historical networks, resulting in the problem of catastrophic forgetting. To tackle such issues, recent advancements in graph neural networks (GNNs) have explored matrix factorization techniques. However, these approaches encounter difficulties in preserving the global patterns of incremental data. In this paper, we propose CLDNE, a Continual Learning framework specifically designed for Dynamic Network Embedding. At the core of CLDNE lies a streaming graph auto-encoder that effectively captures both global and local patterns of the input graph. To further overcome catastrophic forgetting, CLDNE is equipped with an experience replay buffer and a knowledge distillation module, which preserve high-order historical topology and static historical patterns. We conduct experiments on four dynamic networks using link prediction and node classification tasks to evaluate the effectiveness of CLDNE. The outcomes demonstrate that CLDNE successfully mitigates the catastrophic forgetting problem and reduces training time by 80% without a significant loss in learning new patterns.

Enabling Adaptive Sampling for Intra-Window Join: Simultaneously Optimizing Quantity and Quality

>Sampling is one of the most widely employed approximations in big data processing. Among various challenges in sampling design, sampling for join is particularly intriguing yet complex. This perplexing problem starts with a classical case where the join of two Bernoulli samples shrinks its output size quadratically and exhibits a strong dependency on the input data, presenting a unique challenge that necessitates adaptive sampling to guarantee both the quantity and quality of the sampled data. The community has made strides in achieving this goal by constructing offline samples and integrating support from indexes or key frequencies. However, when dealing with stream data, due to the need for real-time processing and high-quality analysis, methods developed for processing static data become unavailable. Consequently, a fundamental question arises: Is it possible to achieve adaptive sampling in stream data without relying on offline techniques? To address this problem, we propose FreeSam, which couples hybrid sampling with intra-window join, a key stream join operator. Our focus lies on two widely used metrics: output size, ensuring quantity, and variance, ensuring quality. FreeSam enables adaptability in both the desired quantity and quality of data sampling by offering control on the two-dimensional space spanned by these metrics. Meanwhile, adjustable trade-offs between quality and performance make FreeSam practical for use. Our experiments show that, for every 1% increase in latency limitation, FreeSam can yield a 3.83% increase in the output size while maintaining the level of the estimator's variance. Additionally, we give FreeSam a multi-core implementation and ensure predictability of its latency through both an analytic model and a neural network model. The accuracy of these models is 88.05% and 96.75% respectively.

Efficient Network-based Fault Detection in Elevator Vibration Signals: A Weighted Fusion Approach of Displacement and Acceleration Data

This research addresses human vibration analysis, rule-based lift fault detection algorithm noise interference, and early fault diagnosis. A new network-based defect diagnosis system uses lift sensor acceleration and displacement data. Convolutional Neural Networks CNNs automatically extract features like motor current (average, maximum, lowest values during a recording session) from this fused data stream and sensor data. CNNs learn key features from pooled data to reduce noise and detect faults early. To build a baseline for normal operation, 132 vibration signal characteristics (RMS, peak-to-peak, mean, standard deviation), lift ID, timestamp, operating mode (Up/Down/Idle), and load condition (Empty/Low/Medium/High) data points are used. Regular operational data and 201 gearbox fault data points are collected. The problem affects motor behaviour via vibration signals, timestamps, lift data, and motor current values. Learning to distinguish this defect type from normal operational data allows the network-based technique identify it. Research shows that network-based strategies outperform traditional methods. It finds weaknesses better than before researches. This method automatically extracts features from the fused data stream to detect defects in real time. Displacement and acceleration data, motor current readings, and CNNs improve noise resistance and failure signal detection. Network-based lift failure detection is advanced in real time problem identification. Precision, early detection, noise resistance, and real-time defect categorization are its strengths. Study affects lift maintenance and find defects early to keep lifts functioning. Real-time defect classification eliminates laborious analysis, simplifying maintenance in which enhancements improve lift performance and reliability.

Ship-based RPA operations for cetacean research in Antarctica: progress, opportunities and challenges

Data collection facilitated by remotely piloted aircraft (RPA) has proven to be revolutionary in many disciplines including for research in extreme environments. Here we assess current use and utility of small multirotor remotely piloted aircraft (RPAs) for the challenging role of facilitating ship-based cetacean research in Antarctica. While such aircraft are now used routinely in sheltered environments in and off Antarctica, a comprehensive literature review found that RPA-mediated cetacean research conducted from ships at sea and outside of the Antarctic Peninsula region was relatively uncommon. In order to determine the potential utility of ship-based multirotor RPA operations for cetacean research, we repeatedly deployed small RPAs during a multidisciplinary research voyage in maritime East Antarctica to collect scientific data contributing to an understanding of krill and krill predator interactions. RPA flight metrics (duration, height, length, speed, distance from ship, battery drainage, satellites acquired) were compared to ship underway environmental sampling data. At a mean duration of 12 minutes, these 139 RPA flights were relatively short yet adequate to achieve the science intended, namely a range of cetacean related data streams including photogrammetry, photo identification, behavioural observations and whale blow sampling in addition to water sampling and collection of general scenic imagery. RPA flight operations were constrained by wind speed but not by air temperature with flights undertaken throughout the full range of air temperatures experienced (down to –9.5°C) but not throughout the full range of wind speeds experienced. For a 12-minute flight duration, battery drainage was around 60% indicating that the RPAs were rarely pushed to their operational limit. There was little evidence that the cold impacted RPA lithium battery performance with estimated maximum flight time within approximately 10% of expected flight time for the RPA platforms most used. Whist small multirotor RPAs are rarely applied to cetacean related research in maritime East Antarctica, we demonstrate their value and potential to deliver data critical to address knowledge gaps that challenge the effective management of both krill and their predators.

Human-Data Interaction as a critical enabler of Electronic Performance Monitoring at construction sites

Human-Data Interaction (HDI) revolves around how humans generate, process, and utilise data. HDI plays a crucial role in evaluating data collection and use in the context of the construction industry, considering the impact on stakeholders such as site managers and labourers. One significant application of HDI is in on-site Electronic Performance Monitoring (EPM), which aims to leverage workplace innovations to enhance productivity, safety, and health. However, the integration and implications of HDI and EPM lack comprehensive understanding. This research seeks to bridge this knowledge gap by presenting a human-data perspective on sensored construction sites, emphasising the challenges and opportunities for driving innovative EPM initiatives. Through a combination of literature review, surveys with HDI experts, and the authors' perspectives and abduction, conceptual frameworks are developed that cluster HDI and EPM. The study's implications are multifaceted, impacting both theoretical understanding and practical applications. The findings highlight the key actors and the data they generate and manipulate across different platforms during EPM deployment. Through the lens of explanatory theories, sociomateriality, and work sociology, the research contributes to understanding the fragmented nature of HDI and EPM as a managerial issue embedded in the work environment. It sheds light on the interactions of actors using digital EPM devices and relevant data streams influenced by the limited agency of specific stakeholders, such as labourers, and the potential neglect of factors related to their well-being. This research distinguishes itself by focusing on the less explored intersection of HDI and EPM in the construction industry. It offers a novel perspective by considering the sensored environment of construction sites as a venue for analysing human-data interactions.

Leveraging State-of-the-Art AI Algorithms in Personalized Oncology: From Transcriptomics to Treatment

Background: Continuous breakthroughs in computational algorithms have positioned AI-based models as some of the most sophisticated technologies in the healthcare system. AI shows dynamic contributions in advancing various medical fields involving data interpretation and monitoring, imaging screening and diagnosis, and treatment response and survival prediction. Despite advances in clinical oncology, more effort must be employed to tailor therapeutic plans based on each patient’s unique transcriptomic profile within the precision/personalized oncology frame. Furthermore, the standard analysis method is not compatible with the comprehensive deciphering of significant data streams, thus precluding the prediction of accurate treatment options. Methodology: We proposed a novel approach that includes obtaining different tumour tissues and preparing RNA samples for comprehensive transcriptomic interpretation using specifically trained, programmed, and optimized AI-based models for extracting large data volumes, refining, and analyzing them. Next, the transcriptomic results will be scanned against an expansive drug library to predict the response of each target to the tested drugs. The obtained target-drug combination/s will be then validated using in vitro and in vivo experimental models. Finally, the best treatment combination option/s will be introduced to the patient. We also provided a comprehensive review discussing AI models’ recent innovations and implementations to aid in molecular diagnosis and treatment planning. Results: The expected transcriptomic analysis generated by the AI-based algorithms will provide an inclusive genomic profile for each patient, containing statistical and bioinformatics analyses, identification of the dysregulated pathways, detection of the targeted genes, and recognition of molecular biomarkers. Subjecting these results to the prediction and pairing AI-based processes will result in statistical graphs presenting each target’s likely response rate to various treatment options. Different in vitro and in vivo investigations will further validate the selection of the target drug/s pairs. Conclusions: Leveraging AI models will provide more rigorous manipulation of large-scale datasets on specific cancer care paths. Such a strategy would shape treatment according to each patient’s demand, thus fortifying the avenue of personalized/precision medicine. Undoubtedly, this will assist in improving the oncology domain and alleviate the burden of clinicians in the coming decade.

State-transition-aware anomaly detection under concept drifts

Detecting temporal abnormal patterns over streaming data is challenging due to volatile data properties and the lack of real-time labels. The abnormal patterns are usually hidden in the temporal context, which cannot be detected by evaluating single points. Furthermore, the normal state evolves over time due to concept drifts. A single model does not fit all data over time. Autoencoders have recently been applied for unsupervised anomaly detection. However, they are trained on a single normal state and usually become invalid after distributional drifts in the data stream. This paper uses an Autoencoder-based approach STAD for anomaly detection under concept drifts. In particular, we propose a state-transition-aware model to map different data distributions in each period of the data stream into states, thereby addressing the model adaptation problem in an interpretable way. In addition, we analyzed statistical tests to detect the drift by examining the sensitivity and powers. Furthermore, we present considerable ways to estimate the probability density function for comparing the distributional similarity for state transitions. Our experiments evaluate the proposed method on synthetic and real-world datasets. While delivering comparable anomaly detection performance as the state-of-the-art approaches, STAD works more efficiently and provides extra interpretability. We also provide insightful analysis of optimal hyperparameters for efficient model training and adaptation.

AOCBLS: A novel active and online learning system for ECG arrhythmia classification with less labeled samples

Electrocardiogram (ECG) is a pivotal determinant of cardiac arrhythmia. In practice, ECG data are often acquired as continuous unlabeled chunks with high labeling costs and severe class imbalances. This poses formidable challenges to existing offline learning methods. To address these issues, we propose a novel model called active online class-specific broad learning system (AOCBLS) to learn streaming imbalanced ECG and classify arrhythmias. First, a two-stage active learning strategy combining diversity and uncertainty queries was designed to select the most valuable beats from the current chunk for further labeling and learning, thereby reducing labeling costs. Subsequently, a stable training weight update rule was developed to address the dynamic imbalances in the streaming data. Finally, an incremental learning algorithm was proposed to update network weights online using selected beats, eliminating the need for retraining from scratch. Thus, the AOCBLS can effectively learn online from imbalanced data streams with less label information. Intra-patient experiments on the benchmark MIT-BIH arrhythmia database showed our model achieved an overall accuracy of 99.10% and a G-mean of 96.35% in identifying the four AAMI-recommended beat types, with only 10% additional annotation, outperforming advanced methods at lower labeling cost. Additionally, patient-specific experimental results on the benchmark database and two other databases, INCARTDB and SVDB, suggest the potential of AOCBLS for personalized ECG monitoring on portable devices.

Real-time anomaly detection of the stochastically excited systems on spherical ([formula omitted]) manifold

Advanced analytical tools have become crucial in today’s constantly changing and complex systems. Real-time Principal Geodesic Analysis (RPGA) is a novel technique that provides a unique perspective for analyzing nonlinear data on differentiable manifolds. Traditional linear methods are often inadequate when exploring the complexities of such data. Orthogonal transformation techniques such as Principal Component Analysis (PCA) and Principal Geodesic Analysis (PGA) are highly desirable for condition monitoring stochastically excited systems in domains like mechanical, aerospace, and civil engineering. However, uncertainties and dynamic fluctuations necessitate robust analytical methods for early change detection to ensure safety, performance, and cost-effectiveness. Limitations posed by linear orthogonal transformation techniques such as PCA and its recursive counterparts make it imperative to adapt these techniques to nonlinear situations where data does not evolve in a flat Euclidean space. Significant advancements have been made in this field over recent decades, with data-driven real-time algorithms such as RPCA, RCCA, and RSSA providing reliable solutions for complex multidimensional problems. However, for curved space, the nonlinear RPGA technique takes center stage. It is known for its effectiveness in extracting meaningful information from the complex data stream. This paper explores the foundational concepts and methodologies underlying the transition from linear to nonlinear data analysis. By examining examples such as stochastic geometric oscillator on S2, and the inverted spherical pendulum cart system navigating a rough surface, we illustrate the significance of reliable, real-time damage detection techniques provided by tools such as RPGA.

Artifact removal from ECG signals using online recursive independent component analysis

The diagnosis of cardiac abnormalities and monitoring of heart health heavily rely on Electrocardiogram (ECG) signals. Unfortunately, these signals frequently encounter interference from diverse artifacts, impeding precise interpretation and analysis. To overcome this challenge, we suggest a novel method for real-time artifact removal from ECG signals through the utilization of Online Recursive Independent Component Analysis (ORICA). Our study outlines a systematic preprocessing pipeline, adaptively estimating the mixing matrix and demixing matrix of the ICA model while streaming data is processed. Additionally, we explore the selection of appropriate ICA components and the use of relevant feature extraction techniques to enhance the quality of extracted cardiac signals. This research presents a promising solution for removing artifacts from ECG signals in real-time, paving the way for improved cardiac diagnostics and monitoring systems. Comparative analyses demonstrate significant improvements in the accuracy of subsequent ECG analysis and interpretation following the application of our ORICA-based preprocessing.

Learning incremental audio–visual representation for continual multimodal understanding

Deep learning methods have demonstrated remarkable success in processing static datasets for various video tasks. However, when confronted with continuous data streams, these approaches often encounter the challenge of catastrophic forgetting. This phenomenon leads to a significant decline in overall performance when learning new classes incrementally. Moreover, existing methods tend to overlook the correlation between audio and visual modalities in video incremental learning, despite their joint significance in scene comprehension. How to continuously learn from new classes while maintaining the knowledge of old videos with limited storage and computing resources is becoming imperative in the field of multimodal learning. In this paper, we introduce CavRL, a pioneering benchmark for audio–visual representation learning under class incremental scenarios. To mitigate catastrophic forgetting, we propose a rehearsal-based training approach that leverages a small exemplar set from previous classes. Our approach constrains the memory buffer within strict storage limits, optimizing exemplar selection by learning correlative audio–visual representations. Additionally, we employ a distillation method to mitigate forgetting in a self-supervised manner. Evaluations on two prevalent multimodal tasks: audio–visual event classification and audio–visual speaker recognition, which demonstrate that CavRL outperforms existing state-of-the-art incremental learning methods across various settings. We anticipate that CavRL will significantly advance research in continual multimodal learning.

Digital ergonomic assessment to enhance the physical resilience of human-centric manufacturing systems in Industry 5.0

The emergence of Industry 5.0 promotes the creation of human-centric values. To fulfill this objective, Internet of Things (IoT) technologies are increasingly being exploited to digitize the human factor and monitor the ergonomics of manual manufacturing systems. These digital assessments, combined with computational algorithms, contribute to the establishment of socially inclusive workplaces while offering detailed insights to safeguard the health of the aging workforce. In this scenario, this study proposes a digital architecture for evaluating the European Assembly Worksheet (EAWS) in human-centric manufacturing systems. Three distinct enabling technologies are leveraged to acquire heterogeneous data streams. A radio-frequency-based smart glove detects the operator’s interactions with the surrounding environment, while a network of marker-less cameras and a four-channel surface Electromyography (sEMG) system capture body joint movements and muscular contractions of the upper limbs, respectively. The acquired data are processed by computational algorithms to define an EAWS-driven set of Key Risk Indicators (KRIs), embedded in an ergonomic decision support system. These risk metrics highlight operator-driven process weaknesses in musculoskeletal, muscular, and material handling dimensions. Finally, the validity of the proposed digital architecture is demonstrated in an industrial-related pilot environment, where an operator assembles a piece of home furniture.

Renewable estimation in expectile regression model with streaming data sets

Streaming data continually expands over time, making it challenging to apply traditional expectile regression methods due to memory limitations. Expectile regression is advantageous for studying the entire conditional distribution of the response to a predictor, but handling streaming data requires innovative approaches. This paper proposes an online renewable expectile regression strategy, which updates the estimator using both the current data and summary statistics from historical data. An ADMM algorithm is employed to efficiently compute the proposed estimator. We establish the consistency and asymptotic normality of this renewable estimator under regular conditions. Extensive simulations and real-data experiments demonstrate that our method performs comparably to the oracle estimator while maintaining computational efficiency and low storage requirements.

Gravitational-Wave Data Analysis with High-Precision Numerical Relativity Simulations of Boson Star Mergers.

Gravitational-wave signals detected to date are commonly interpreted under the paradigm that they originate from pairs of black holes or neutron stars. Here, we explore the alternative scenario of boson-star signals being present in the data stream. We perform accurate and long (∼20 orbits) numerical simulations of boson-star binaries and inject the resulting strain into LIGO noise. Our Bayesian inference reveals that some boson-star signals are degenerate with current approximants, albeit with biased parameters, while others exhibit smoking-gun signatures leaving behind conspicuous residuals.

Universal all-optical zero-bias logic gates in silicon photonics

Universal all-optical (optical input – optical output) logic gates, realized in a commercial foundry silicon photonics process, are reported. Microring resonator modulator and zero-biased stacked photodiodes are used to create the required optical input-output nonlinear transfer function for optical gate realizations. Several logic gate functions, specifically NOR which is a universal logic gate that can be used to implement any logic, are demonstrated. Logic operation on 10 Mbps data streams is demonstrated using universal optical logic gates without requiring any electrical supply voltage. The total optical power required for one such universal logic gate is 160 µW.

Floor-Level Occupancy Estimation of a Multi-Story Building Using Coarse Wi-Fi Data

In recent years, there has been an extraordinary increase in wireless capable devices and network infrastructure, which spawned a corresponding rise in data produced from the interactions of these technologies. Mobile devices constantly roam, leading to a perpetual dialog between a mobile device and wireless access points. This dialogue generates a continuous stream of device-specific data, including but not limited to a device’s media access control address, time of access, and received signal strength. Given the knowledge of the access point’s location and received signal strength, it is possible to infer the position of user devices and estimate their mobility and occupancy. This article presents two methods for accurately measuring floor-level occupancy in a multi-story building at Texas State University using coarse Wi-Fi log data. The first method employs a static filter, while the second incorporates user-role data and user location to create a dynamic filter. Quantitative methods are used to evaluate these filters against field-collected reference data and existing internal people-counting sensors. Our results demonstrate that the dynamic filter, leveraging variable thresholds, provides a more accurate estimation of occupancy compared to the fixed 5-min static filter which consistently overestimated occupancy. This research sheds light on the potential of dynamic filters derived from user-role data for precise floor-level occupancy estimations, with implications for various applications.

EdgeSyn: Privacy-preserving Data Publishing on Edge Network over Infinite Multimedia Data Stream

To privately publish sensitive multimedia data in an edge network with fog devices, one of the best privacy-preserving solutions is to use differential privacy (DP) mechanisms. However, existing DP data publication mechanisms for the infinite data stream of edge networks mainly focus on publishing data with specific types of data or a set of predetermined queries. This approach is not suitable for multimedia data with numerous features that require a more flexible data publishing mechanism. In this paper, we propose EdgeSyn, a novel mechanism for accurately publishing multimedia data over infinite data streams in an edge network. It allocates privacy budgets with a sliding window, adopting data synthesis mechanisms to support dynamic publishing without loss of accuracy. In more detail, EdgeSyn addresses the limitations associated with data types in prior data stream publishing approaches and introduces a privacy budget management strategy that optimally allocates budgets for the implementation of data synthesis mechanisms over an infinite data stream. The experimental results show that EdgeSyn performs well under different privacy budgets and various lengths of active windows.

Unsupervised anomaly detection of multivariate time series based on multi-standard fusion

With the wide application of the Internet of Things, industrial systems generate a large amount of multivariate time series data every day. Industrial data often contains a wide variety of anomalies. The existing anomaly detection methods have some shortcomings in dealing with large-scale, high-dimensional, and real-time industrial data. Particularly in the absence of prior knowledge, unsupervised learning methods are easily influenced by noise. Therefore, we propose a new unsupervised anomaly detection algorithm of multivariate time series based on multi-standard fusion (MSAD). Specifically, MSAD first classifies the data by analyzing the data density and sample spacing, thereby converting unsupervised anomaly detection into weakly supervised anomaly detection. Then, based on the characteristics of the data and deep learning technology, the degree of abnormality in the sample is comprehensively measured from four perspectives: holistic, local, time, and deep feature learning. At this stage, MSAD 's ability to learn edge features of data has been improved. Finally, based on MSAD, we design an anomaly detection method for industrial data streams in combination with data filtering techniques. The experimental results show that MSAD can detect anomalies more accurately and show stronger robustness to the influence of noise.