Dynamic Datasets Research Articles

Methane Emissions From Land and Aquatic Ecosystems in Western Siberia: An Analysis With Methane Biogeochemistry Models

AbstractWestern Siberia contains extensive wetlands and aquatic ecosystems, contributing a significant amount of methane (CH4) emissions to the atmosphere. However, estimates of CH4 fluxes over the region are poorly constrained partly due to the uncertainties from the inundated area data. This study applied two process‐based biogeochemistry models to quantify the emissions from land and aquatic ecosystems over the region within the period 2000–2021 using different inundation datasets. To drive land methane modeling, we use one static wetland map and one dynamic wetland area data set called Wetland Area and Dynamics for Methane Modeling (WAD2M) (2000–2020). To drive lake methane modeling, we use the surface area of aquatic ecosystems from three datasets: (a) HydroLAKES; (b) Global Surface Water (GSW); and (c) surface water inundation from Soil Moisture Active Passive (SMAP) (2016–2021). Using these datasets, we conduct four simulations to compare emissions over the region. We find that the net methane emissions from land using the static wetland map are larger than those using WAD2M. SMAP and GSW estimate larger emissions than HydroLAKES does from aquatic ecosystems. Total emissions over the region range from 4.80 ± 0.43 to 8.29 ± 0.81 Tg CH4/year from 2016 to 2020, which is the intersection period of four simulations. This study is among the first to investigate methane emissions from the whole landscape in the region. Our study highlights the importance of dynamic wetland and aquatic area data in quantifying regional methane emissions.

Optimization of painting efficiency applying unique techniques of high-voltage conductors and nitrotherm spray: Developing deep learning models using computational fluid dynamics dataset

The impetus of the current three-dimensional Eulerian–Lagrangian work is to analyze the impact of simultaneously using the inventive high-voltage conductors and Nitrotherm spraying technique for maximizing the industrial painting process efficiency. This investigation employs high-fidelity computational fluid dynamics (CFD) results in deep learning models as an input dataset. The novel conductors are called high-voltage retractable blades (HVRB) and high-voltage adjustable control-ring (HVACR) mounted on the head of the electrostatic rotating bell sprayer. The influence of dominant operational parameters, such as temperature and velocity of injected nitrogen or air, droplets' electric charge values, and their size ranges, and electric field density are examined in the considered database for the Nitrotherm spraying methodology. This broad range of parametric investigation illustrates that the inclusion of shaping nitrogen flow, manipulated electric field density, and droplet charging weights significantly affect the spraying deposition rate. The pressurized clean heated nitrogen flow, which is injected from the nozzles of the atomizers, positively redirects and harmonizes the charged droplets that construct an optimized spray plume pattern with a smaller diameter. Using innovative HVRB and HVACR conductors is manipulated the electric fields and leads to denser distribution, intensifying the acting electric force on the droplets, resulting in higher spraying transfer efficiency (TE) and thicker film formation. Based on the results, employing the introduced conductors in combination with the heated nitrogen instead of air leads to higher TE, rare overspray occurrence, formation of an esthetic paint film, lower paint consumption, and application time. Also, the collected complete database is employed for machine learning investigation to predict flow with high accuracy, aiming to reduce computational time/cost. A convolutional auto-encoder is used to reduce the computational cost with just 10% of the initial CFD computations, with a mean error of 1% on the prediction of the deposited droplet areas of the spray. The analysis revealed that by employing recurrent convolutional layers, superior capturing of the input pattern is obtained, which significantly aids the final prediction.

Social learning and the complex contagion of political symbols in Twitter: The case of the yellow ribbon in Catalonia

In this paper, we empirically test whether the spread of political symbols in Twitter is due to complex contagion. We analyzed behavior consisting of editing the Twitter account name to include an icon with a yellow ribbon; a symbol that represents the demand for the release of imprisoned Catalan politicians and civil leaders. To test this hypothesis, we used a behavioral, non-reactive, relational, and dynamic dataset of a large sample of potential users. First, we show that the probability of displaying a ribbon is associated with the proportion of peers who also display it (friends that share their support for the political cause). Second, we rule out alternative explanations as simple contagion and homophily. To rule out simple contagion, we run three empirically calibrated, agent-based simulations. We use our dataset to rule out homophily. And third, we suggest that adoption cannot be interpreted as the result of a compliance mechanism or as the result of normative pressures. Instead, the most plausible micro-level generative mechanism that leads to a complex contagion pattern is a peer learning process. Our study makes several contributions to the field. We show how digital data can be effectively used to identify new explananda and test the plausibility of competing behavioral explanatory mechanisms. We also contribute to the development of the theory of complex contagions. Our study widens the set of conditions for complex contagion and the set of reasons to explain why complex contagion might occur.

Incremental reduction methods based on granular ball neighborhood rough sets and attribute grouping

In the framework of rough sets, incremental algorithms can effectively reduce repetitive computations for dynamic datasets by discovering the update principles of relevant knowledge. Considering the variations in attribute sets, incremental methods based on neighborhood granulation provide fast updates to attribute reduction in continuous-valued information systems. However, most of these methods assume that the sample distribution and neighborhood radius remain fixed with changes in the attribute set, which may mislead the granulation process and affect the model's classification ability. In this paper, a new incremental reduction method based on granular balls and attribute grouping is proposed for dynamic information systems with multiple attribute additions. Different neighborhood radii are adaptively determined when the attribute set changes, and the number of neighborhood granules can also be effectively reduced. Furthermore, the attributes are grouped based on the k-means algorithm, and only attributes from different groups or with small relevance to those in the current reduction set are considered to be incorporated as a reduction attribute, thus reducing the computation time and simultaneously preserving informative attributes. Linear and nonlinear correlation coefficients are used to measure attribute relevance based on which three corresponding incremental reduction algorithms are developed. Finally, extensive experimental results on 12 benchmark datasets are shown to compare the proposed method with non-incremental and typical incremental methods in terms of time cost, classification accuracy, the size of attribute reduction, and coverage of the reduction.

Fraud Detection of Credit Cards Using Supervised Machine Learning Techniques

Credit card fraud encompasses illicit activities aimed at unlawfully obtaining confidential information to enable unauthorized individuals to engage in illegal transactions. As technology advances, fraudsters have honed their skills in evading security measures, presenting a formidable challenge in fraud detection. To address this issue, an array of algorithms and analytical techniques has emerged to identify and mitigate instances of fraud. This research aimed to identify the most appropriate supervised machine learning algorithm for credit card fraud detection. Logistic Regression, Random Forest, Support Vector Machine, and Decision Trees were implemented and compared. Due to the imbalanced nature of the dataset, the SMOTE (Synthetic Minority Oversampling Technique) technique was employed to rectify the data imbalance by oversampling the minority class. The performance of the trained models was evaluated using various metrics, including the confusion matrix, accuracy, precision, recall, f1-score, Matthews Correlation Coefficient (MCC), and Area Under the Curve (AUC). The results of the analysis revealed that Random Forests exhibited exceptional performance, achieving an impressive recall score of 84% and surpassing other algorithms. This research provides the groundwork for future investigations involving diverse deep-learning techniques applied to real-time and dynamic datasets, enabling continuous enhancements in fraud detection and prevention mechanisms.

Fast and accurate semantic segmentation of road crack video in a complex dynamic environment

ABSTRACT This paper proposes a fast and accurate semantic segmentation of road crack video in a complex dynamic environment. First, a fast key frame selection algorithm is designed by combining the interframe dissimilarity constrained video frame difference method (FDM) and shot interval sampling method (SISM). Second, the complex and dynamic characteristics of application scenarios are studied, and a high-precision crack semantic segmentation DBPNet containing a densely diverse branch module (DDBM) and diverse branch pyramid module (DBPM) is proposed, which can not only focus on the local aggregation of the same scale feature but also fuse and reconstruct long-distance related semantic features. Finally, a dynamic video data set of cracks in complex environments, including multiple types of interference, different weather and lighting is established, and experimental analysis is carried out. The results show that the proposed method can improve the efficiency of video crack detection by 5–7 times compared to the method without using key frames, and the detection accuracy can reach 72.85%, which can adapt to dynamic video changes in a variety of complex scenes. The proposed network, DBPNet, outperforms the current state-of-the-art methods in road crack semantic segmentation on challenging data sets in terms of both Dice and MIOU.

Self-Paced Learning Based Graph Convolutional Neural Network for Mixed Integer Programming (Student Abstract)

Graph convolutional neural network (GCN) based methods have achieved noticeable performance in solving mixed integer programming problems (MIPs). However, the generalization of existing work is limited due to the problem structure. This paper proposes a self-paced learning (SPL) based GCN network (SPGCN) with curriculum learning (CL) to make the utmost of samples. SPGCN employs a GCN model to imitate the branching variable selection during the branch and bound process, while the training process is conducted in a self-paced fashion. Specifically, SPGCN contains a loss-based automatic difficulty measurer, where the training loss of the sample represents the difficulty level. In each iteration, a dynamic training dataset is constructed according to the difficulty level for GCN model training. Experiments on four NP-hard datasets verify that CL can lead to generalization improvement and convergence speedup in solving MIPs, where SPL performs better than predefined CL methods.

Dynamic Heterogeneous Graph Attention Neural Architecture Search

Dynamic heterogeneous graph neural networks (DHGNNs) have been shown to be effective in handling the ubiquitous dynamic heterogeneous graphs. However, the existing DHGNNs are hand-designed, requiring extensive human efforts and failing to adapt to diverse dynamic heterogeneous graph scenarios. In this paper, we propose to automate the design of DHGNN, which faces two major challenges: 1) how to design the search space to jointly consider the spatial-temporal dependencies and heterogeneous interactions in graphs; 2) how to design an efficient search algorithm in the potentially large and complex search space. To tackle these challenges, we propose a novel Dynamic Heterogeneous Graph Attention Search (DHGAS) method. Our proposed method can automatically discover the optimal DHGNN architecture and adapt to various dynamic heterogeneous graph scenarios without human guidance. In particular, we first propose a unified dynamic heterogeneous graph attention (DHGA) framework, which enables each node to jointly attend its heterogeneous and dynamic neighbors. Based on the framework, we design a localization space to determine where the attention should be applied and a parameterization space to determine how the attention should be parameterized. Lastly, we design a multi-stage differentiable search algorithm to efficiently explore the search space. Extensive experiments on real-world dynamic heterogeneous graph datasets demonstrate that our proposed method significantly outperforms state-of-the-art baselines for tasks including link prediction, node classification and node regression. To the best of our knowledge, DHGAS is the first dynamic heterogeneous graph neural architecture search method.

Intensity-Aware Loss for Dynamic Facial Expression Recognition in the Wild

Compared with the image-based static facial expression recognition (SFER) task, the dynamic facial expression recognition (DFER) task based on video sequences is closer to the natural expression recognition scene. However, DFER is often more challenging. One of the main reasons is that video sequences often contain frames with different expression intensities, especially for the facial expressions in the real-world scenarios, while the images in SFER frequently present uniform and high expression intensities. Nevertheless, if the expressions with different intensities are treated equally, the features learned by the networks will have large intra-class and small inter-class differences, which are harmful to DFER. To tackle this problem, we propose the global convolution-attention block (GCA) to rescale the channels of the feature maps. In addition, we introduce the intensity-aware loss (IAL) in the training process to help the network distinguish the samples with relatively low expression intensities. Experiments on two in-the-wild dynamic facial expression datasets (i.e., DFEW and FERV39k) indicate that our method outperforms the state-of-the-art DFER approaches. The source code will be available at https://github.com/muse1998/IAL-for-Facial-Expression-Recognition.

UNDERSTANDING THE CHALLENGES AROUND DESIGN ACTIVITIES THAT INCORPORATE BEHAVIORAL DATA

AbstractRecently, methods and approaches such as Participatory Data Analysis, Data-Enabled Design, and Contextual Inquiry have highlighted how design activities can benefit from behavioral data. This data offers new ways to learn from what people do and how they do it, across time and space. However, behavioral data introduces changes and frictions to design activities and poses several challenges for designers to overcome. In this paper, we conduct two workshops with 18 expert designers, from industry and academia, to understand the nature of these challenges, beyond the technical aspects. We contribute by underlining the challenges and opportunities of incorporating behavioral data into design activities; including a design perspective on data, interacting with participants, and interacting with regulatory bodies. We translate our findings into opportunities for a better alignment between regulatory bodies, designers, and participants. We propose to harness the iterative nature of design activities and embedded it into a process that allows for continuous reflection, reassessment, and review of highly dynamic datasets.

A Hybrid Continuous-Time Dynamic Graph Representation Learning Model by Exploring Both Temporal and Repetitive Information

Recently, dynamic graph representation learning has attracted more and more attention from both academic and industrial communities due to its capabilities of capturing different real-world phenomena. For a dynamic graph represented as a sequence of timestamped events, there are two kinds of evolutionary essences: temporal and repetitive information. At present, the temporal information of interactions (e.g., timestamps) have been deeply explored. However, as another vital nature of dynamic graphs, the repetitive information of interactions between two nodes is neglected, which may lead to inaccurate node representation. To address this issue, we propose a novel continuous-time dynamic graph representation learning model, which consists of a node-level-memory based module, a historical high-order neighborhood based vertical aggregation module and a repetitive-topological information based horizontal aggregation module. In particular, to characterize the evolving pattern of the repetitive information of interactions between a pair of nodes, we put forward a repetitive-interaction based attention mechanism to integrate the two key attributes (i.e., the content and the number of interactions) of repetitive interactions at different moments, based on the insight that the repetitive behaviors of nodes are widespread and essential. We conduct extensive experiments including future link prediction tasks (for transductive and inductive learning) and dynamic node classification task, and results on three real-life dynamic graph datasets demonstrate that the proposed method significantly outperforms state-of-the-art baselines, for both observed nodes and new ones.

GeaFlow: A Graph Extended and Accelerated Dataflow System

GeaFlow is a distributed dataflow system optimized for streaming graph processing, and has been widely adopted at Ant Group, serving various scenarios ranging from risk control of financial activities to analytics on social networks and knowledge graphs. It is built on top of a base with full-fledged stateful stream processing capabilities, extended with a series of graph-aware optimizations to address the space explosion and programming complexity issues of conventional join-based approaches. We propose new state backends and streaming operators that facilitate processing on dynamic graph-structured datasets, reducing space consumed by states. We develop a hybrid domain-specific language that embeds Gremlin into SQL, supporting both table and graph abstractions over streaming data. In addition to streaming workloads, GeaFlow is also extensively used for some batch processing jobs. In the largest deployments to date, GeaFlow is able to process tens of millions of events per second and manage hundreds of terabytes of states.

Cardiac phase detection in echocardiography using convolutional neural networks

Echocardiography is a commonly used and cost-effective test to assess heart conditions. During the test, cardiologists and technicians observe two cardiac phases—end-systolic (ES) and end-diastolic (ED)—which are critical for calculating heart chamber size and ejection fraction. However, non-essential frames called Non-ESED frames may appear between these phases. Currently, technicians or cardiologists manually detect these phases, which is time-consuming and prone to errors. To address this, an automated and efficient technique is needed to accurately detect cardiac phases and minimize diagnostic errors. In this paper, we propose a deep learning model called DeepPhase to assist cardiology personnel. Our convolutional neural network (CNN) learns from echocardiography images to identify the ES, ED, and Non-ESED phases without the need for left ventricle segmentation or electrocardiograms. We evaluate our model on three echocardiography image datasets, including the CAMUS dataset, the EchoNet Dynamic dataset, and a new dataset we collected from a cardiac hospital (CardiacPhase). Our model outperforms existing techniques, achieving 0.96 and 0.82 area under the curve (AUC) on the CAMUS and CardiacPhase datasets, respectively. We also propose a novel cropping technique to enhance the model’s performance and ensure its relevance to real-world scenarios for ES, ED, and Non ES-ED classification.

Economic growth and poverty reduction: Evidence from Sub‐Saharan African (SSA) countries

AbstractSub‐Saharan African (SSA) countries have registered outstanding economic growth in recent decades. However, poverty is still pervasive, deep, and severe in the region. This paper aims to analyze the impact of the recent economic growth on poverty reduction in 27 SSA countries using a new dynamic panel data set. Adopting System GMM estimation, this study found that economic growth has been associated with poverty reduction. Also, the previous level of poverty had an intense impact on the prevalent poverty status in the region. Moreover, although nonmonetary measures and income measures of poverty produce similar results (showing a reduction in poverty), nonincome poverty, especially destitution data, suggest that SSA countries are poorer even than previously understood. We suggest that SSA countries should promote a policy of income‐enhancing (i.e., economic growth) to ramp up poverty reduction.

Automatic Multiple Articulator Segmentation in Dynamic Speech MRI Using a Protocol Adaptive Stacked Transfer Learning U-NET Model.

Dynamic magnetic resonance imaging has emerged as a powerful modality for investigating upper-airway function during speech production. Analyzing the changes in the vocal tract airspace, including the position of soft-tissue articulators (e.g., the tongue and velum), enhances our understanding of speech production. The advent of various fast speech MRI protocols based on sparse sampling and constrained reconstruction has led to the creation of dynamic speech MRI datasets on the order of 80-100 image frames/second. In this paper, we propose a stacked transfer learning U-NET model to segment the deforming vocal tract in 2D mid-sagittal slices of dynamic speech MRI. Our approach leverages (a) low- and mid-level features and (b) high-level features. The low- and mid-level features are derived from models pre-trained on labeled open-source brain tumor MR and lung CT datasets, and an in-house airway labeled dataset. The high-level features are derived from labeled protocol-specific MR images. The applicability of our approach to segmenting dynamic datasets is demonstrated in data acquired from three fast speech MRI protocols: Protocol 1: 3 T-based radial acquisition scheme coupled with a non-linear temporal regularizer, where speakers were producing French speech tokens; Protocol 2: 1.5 T-based uniform density spiral acquisition scheme coupled with a temporal finite difference (FD) sparsity regularization, where speakers were producing fluent speech tokens in English, and Protocol 3: 3 T-based variable density spiral acquisition scheme coupled with manifold regularization, where speakers were producing various speech tokens from the International Phonetic Alphabetic (IPA). Segments from our approach were compared to those from an expert human user (a vocologist), and the conventional U-NET model without transfer learning. Segmentations from a second expert human user (a radiologist) were used as ground truth. Evaluations were performed using the quantitative DICE similarity metric, the Hausdorff distance metric, and segmentation count metric. This approach was successfully adapted to different speech MRI protocols with only a handful of protocol-specific images (e.g., of the order of 20 images), and provided accurate segmentations similar to those of an expert human.

Synchronous spatio-temporal signature verification via Fusion Triplet Supervised Network

Handwritten signature has been regarded as one of the most accepted and practical means of person verification since ancient times. Unlike existing methods to trace the signer only considering a single spatial or temporal information, in this paper, we propose to combine two forms of signature data (stroke images and sensor signals) to realize person verification using the novel Fusion Triplet Supervised Network (fuseTSN). First, the Electronic Handwritten Signature Acquisition System 1.0 (EHS-AS 1.0) connected to the Wacom tablet is developed to construct a synchronous static–dynamic Chinese signature dataset (SynCS) and to transform existing dynamic signature datasets into static–dynamic datasets. Then, the hybrid feature extractor (ResCNN-BiLSTM with attention) as the backbone is embedded into the fuseTSN model, mapping the images and time-series signal into a common hyperspace. Furthermore, the similarity of the input [Anchor, Positive, Test] is measured to produce the final output. Finally, the proposed fuseTSN is tested on three public datasets and also evaluated on our SynCS dataset comprised of 24000 unique triplet label samples. The results show that the accuracy of the fuseTSN model is at least 5.01% higher than that of a unimodal system with only stroke image or sensor data on the SynCS dataset and outperforms the current state-of-the-art multimodal methods for writer-independence (WI) verification on public datasets. In addition, we present the fuseTSN’s reliable judgment basis for spatial strokes in the heat-map and further find that the combination of time attributes (OA and OE) can significantly improve the model performance.

Self-supervised Dynamic Graph Embedding with evolutionary neighborhood and community

Real-world data and systems are generally evolutionary over time, which can be formalized as dynamic graphs, and Dynamic Graph Embedding (DGE) as a fundamental technique for analyzing dynamic graphs, has shown considerable potential in recent years. However, the majority of existing DGEs focus only on the local neighborhood structure, ignoring the global community structure, which is also an important property of dynamic graphs. Therefore, how to fuse both local and global information to learn more comprehensive graph representations and their variation patterns is a meaningful and promising topic. In this paper, we propose a novel DGE method to explore the graph properties and evolution patterns from Neighborhood and Community views, named DGNC. The neighborhood view learns the node embeddings in line with timestamped interaction events, and models the local variation of graphs via the time-varying attention vector. While the community view encodes the alternative transformation of node features and community features according to community correlation, which aims to extract the global evolution of graphs. To better integrate the two views, cross-supervised contrastive learning is introduced to enable them to collaboratively provide guidance and complementary information to each other, which can significantly improve the quality of graph embedding. We evaluate DGNC through diverse downstream tasks on four real-world dynamic graph datasets, and the experimental results verify the effectiveness and superiority of our proposed method.

MONITORING DATA AGGREGATION OF DYNAMIC SYSTEMS USING INFORMATION TECHNOLOGIES

The subject matter of the article is models, methods and information technologies of monitoring data aggregation. The goal of the article is to determine the best deep learning model for reducing the dimensionality of dynamic systems monitoring data. The following tasks were solved: analysis of existing dimensionality reduction approaches, description of the general architecture of vanilla and variational autoencoders, development of their architecture, development of software for training and testing of autoencoders, conducting research on the performance quality of autoencoders for the problem of dimensionality reduction. The following models and methods were used: data processing and preparation, data dimensionality reduction. The software was developed using the Python language. Scikit-learn, Pandas, PyTorch, NumPy, argparse and others were used as auxiliary libraries. Obtained results: the work presents a classification of models and methods for dimensionality reduction, general reviews of vanilla and variational autoencoders, which include a description of the models, their properties, loss functions and their application to the problem of dimensionality reduction. Custom autoencoder architectures were also created, including visual representations of the autoencoder architecture and descriptions of each component. The software for training and testing autoencoders was developed, the dynamic system monitoring data set, and the steps for pre-training the data set were described. The metric for evaluating the quality of models is also described; the configuration of autoencoders and their training are considered. Conclusions: The vanilla autoencoder recovers the data much better than the variational one. Looking at the fact that the architectures of the autoencoders are the same, except for the peculiarities of the autoencoders, it can be noted that a vanilla autoencoder compresses data better by keeping more useful variables for later recovery from the bottleneck. Additionally, by training on different bottleneck sizes, you can determine the size at which the data is recovered best, which means that the most important variables are preserved. Looking at the results in general, the autoencoders work effectively for the dimensionality reduction task and the data recovery quality metric shows that they recover the data well with an error of 3–4 digits after 0. In conclusion, the vanilla autoencoder is the best deep learning model for aggregating monitoring data of dynamic systems.

SVD-SLAM: Stereo Visual SLAM Algorithm Based on Dynamic Feature Filtering for Autonomous Driving

The conventional Simultaneous Localization and Mapping (SLAM) algorithm assumes a static world, which is easily influenced by dynamic elements of the surrounding environment. For high-precision localization in dynamic scenes, a dynamic SLAM algorithm combining instance segmentation and dynamic feature point filtering is proposed to address this issue. Initially, YOLACT-dyna, a one-stage instance segmentation network, was developed in order to perform instance segmentation on the input image, eliminate potential moving objects in the scene, and estimate the camera pose roughly. Second, based on the camera pose and polar constraint, the motion probability of each possible moving object was computed. Finally, the moving feature points were filtered out, and the static feature points were used to calculate the pose. The experimental results reveal that this algorithm’s recall rate in the dynamic regional KITTI dataset was 94.5% in public datasets. Accuracy is enhanced in environments with dynamic object location. At the same time, it can guarantee the positioning accuracy of a static scene, effectively enhancing the visual SLAM system’s position precision and robustness in a dynamic environment. It can meet the requirements of the automatic driving system’s real-time operation.

Epileptic Seizure Classification and Feature Optimization Technique Using Grey Wolf Algorithm on Dynamic Datasets

Epileptic Seizure Classification and Feature Optimization Technique Using Grey Wolf Algorithm on Dynamic Datasets