Electrocardiography (ECG) is the gold standard for monitoring heart function and is crucial for preventing the worsening of cardiovascular diseases (CVDs). However, the inconvenience of ECG acquisition poses challenges for long-term continuous monitoring. Consequently, researchers have explored non-invasive and easily accessible photoplethysmography (PPG) as an alternative, converting it into ECG. Previous studies have focused on peaks or simple mapping to generate ECG, ignoring the inherent periodicity of cardiovascular signals. This results in an inability to accurately extract physiological information during the cycle, thus compromising the generated ECG signals' clinical utility. To this end, we introduce a novel PPG-to-ECG translation model called CATransformer, capable of adaptive modeling based on the cardiac cycle. Specifically, CATransformer automatically extracts the cycle using a cycle-aware module and creates multiple semantic views of the cardiac cycle. It leverages a transformer to capture detailed features within each cycle and the dynamics across cycles. Our method outperforms existing approaches, exhibiting the lowest RMSE across five paired PPG-ECG databases. Additionally, extensive experiments are conducted on four cardiovascular-related tasks to assess the clinical utility of the generated ECG, achieving consistent state-of-the-art performance. Experimental results confirm that CATransformer generates highly faithful ECG signals while preserving their physiological characteristics.

The rapid increase in remote sensing satellites has led to the emergence of distributed space-based observation systems. However, existing distributed remote sensing models often rely on centralized training, resulting in data leakage, communication overhead, and reduced accuracy due to data distribution discrepancies across platforms. To address these challenges, we propose theSelf-Adjustment FEderated Learning(SAFE) framework, which innovatively leverages federated learning to enhance collaborative sensing in remote sensing scenarios. SAFE introduces four key strategies: (1)Class Rectification Optimization, which autonomously addresses class imbalance under unknown local and global distributions. (2)Feature Alignment Update, which mitigates Non-IID data issues via locally controlled EMA updates. (3)Dual-Factor Modulation Rheostat, which dynamically balances optimization effects during training. (4)Adaptive Context Enhancement, which is designed to improve model performance by dynamically refining foreground regions, ensuring computational efficiency with accuracy improvement across distributed satellites. Experiments on real-world image classification and object segmentation datasets validate the effectiveness and reliability of the SAFE framework in complex remote sensing scenarios.

Process-aware information systems (PAIS) are designed to manage and support business processes within an organization. Drift detection in process mining aims to detect process changes by analyzing event logs, which can guarantee the accuracy and reliability of business processes within PAIS. However, existing methods often only consider process changes from a single perspective (e.g., control-flow), without fully utilizing other information in event logs like resources and timestamps. In this paper, we propose a novel Multi-view Drift Detection method, called MDD, to detect changes with graph convolutional networks. Specifically, 1) multiple k-nearest neighbors graphs are constructed for two log segments based on multiple perspectives, and graph convolutional networks are employed to capture features from these graphs. Meanwhile, we utilize a fully connected network to fuse multiple knn graphs, generating two fusion graphs; 2) we use Wasserstein distance to measure the distance between two fusion graphs after feature fusion; 3) the K-means algorithm is adopted to find the candidate drift points, and then the actual change points will be identified with a filter mechanism. Experimental results demonstrate that the MDD effectively identifies drift points on simulated logs and real-life logs.

We introduce a novel method tailored for unconstrained multiview optical satellite photogrammetry in time-varying illumination and reflection conditions. Our approach uses continuous radiance fields to represent surface radiance and albedo based on radiometry principles, integrating both static and transient components for satellite photogrammetry. In addition, an innovative self-supervised mechanism is introduced to optimize the learning process which leverages dark regions’ accentuation, transient and static composition, and shadow regularization. Evaluations on multidate WorldView-3 images affirm that our model consistently surpasses the state-of-the-art techniques.

Throughout human history, deadly infectious diseases emerged occasionally. Even with the present-day advanced healthcare systems, the COVID-19 has caused more than six million deaths worldwide (as of 27 July 2022). Currently, researchers are working to develop tools for better and effective management of the pandemic. “Contact tracing” is one such tool to monitor and control the spread of the disease. However, manual contact tracing is labor-intensive and time-consuming. Therefore, manually tracking all potentially infected individuals is a great challenge, especially for an infectious disease like COVID-19. To date, many digital contact tracing applications were developed and used globally to restrain the spread of COVID-19. In this work, we perform a detailed review of the current digital contact tracing technologies. We mention some of their key limitations and propose a fully integrated system for contact tracing of infectious diseases using COVID-19 as a case study. Our system has four main modules—1) case maps; 2) exposure detection; 3) screening; and 4) health indicators that take multiple inputs like users’ self-reported information, measurement of physiological parameters, and information of the confirmed cases from the public health, and keeps a record of contact histories using Bluetooth technology. The system can potentially evaluate the users’ risk of getting infected and generate notifications to alert them about the exposure events, risk of infection, or abnormal health indicators. The system further integrates the Web-based information on confirmed COVID-19 cases and screening tools, which potentially increases the adoption rate of the system.

Unmanned aerial vehicles (UAVs) have been used extensively in traffic data collection owing to their flexibility, stability, and ease of operation. However, vehicle detection methods using horizontal detectors are less efficient and accurate for UAV vehicle detection, whereas rotated detectors suffer the discontinuous boundaries problem. Therefore, we proposed a traffic data extraction framework based on YOLOv5-OBB object detection and DeepSORT-OBB tracking algorithms to extract highly accurate traffic data from UAV videos. The framework was tested using aerial videos recorded by a UAV-mounted high-definition camera. Field experiments were conducted to collect reference data from an onboard high-precision sensor for use in evaluating the precision of the extracted traffic data. The traffic data, including trajectory, yaw angle, speed, and heading of vehicles, were extracted from UAV videos captured in different traffic scenes. The overall extraction accuracy reached 98.5%, indicating the reliability of the proposed framework in extracting highly accurate traffic data.

Low-resource Relation Extraction (LRE) aims to extract relation facts from limited labeled corpora when human annotation is scarce.Existing works either utilize self-training scheme to generate pseudo labels that will cause the gradual drift problem, or leverage metalearning scheme which does not solicit feedback explicitly.To alleviate selection bias due to the lack of feedback loops in existing LRE learning paradigms, we developed a Gradient Imitation Reinforcement Learning method to encourage pseudo label data to imitate the gradient descent direction on labeled data and bootstrap its optimization capability through trial and error.We also propose a framework called GradLRE, which handles two major scenarios in low-resource relation extraction.Besides the scenario where unlabeled data is sufficient, GradLRE handles the situation where no unlabeled data is available, by exploiting a contextualized augmentation method to generate data.Experimental results on two public datasets demonstrate the effectiveness of GradLRE on low resource relation extraction when comparing with baselines.Source code is available 1 .

atural language dialogue systems raise great attention recently. As many dialogue models are data-driven, high-quality datasets are essential to these systems. In this paper, we introduce Pchatbot, a large-scale dialogue dataset that contains two subsets collected from Weibo and Judicial forums respectively. To adapt the raw dataset to dialogue systems, we elaborately normalize the raw dataset via processes such as anonymization, deduplication, segmentation, and filtering. The scale of Pchatbot is significantly larger than existing Chinese datasets, which might benefit the data-driven models. Besides, current dialogue datasets for personalized chatbot usually contain several persona sentences or attributes. Different from existing datasets, Pchatbot provides anonymized user IDs and timestamps for both posts and responses. This enables the development of personalized dialogue models that directly learn implicit user personality from the user's dialogue history. Our preliminary experimental study benchmarks several state-of-the-art dialogue models to provide a comparison for future work. The dataset can be publicly accessed at Github: https://github.com/qhjqhj00/Pchatbot.

The real-world networks often compose of different types of nodes and edges with rich semantics, widely known as heterogeneous information network (HIN). Heterogeneous network embedding aims to embed nodes into low-dimensional vectors which capture rich intrinsic information of heterogeneous networks. However, existing models either depend on manually designing meta-paths, ignore mutual effects between different semantics, or omit some aspects of information from global networks. To address these limitations, we propose a novel Graph-Aggregated Heterogeneous Network Embedding (GAHNE), which is designed to extract the semantics of HINs as comprehensively as possible to improve the results of downstream tasks based on graph convolutional neural networks. In GAHNE model, we develop several mechanisms that can aggregate semantic representations from different single-type sub-networks as well as fuse the global information into final embeddings. Extensive experiments on three real-world HIN datasets show that our proposed model consistently outperforms the existing state-of-the-art methods.

We explore how to leverage the performance of face feature points detection on mobile terminals from 3 aspects. First, we optimize the models used in SDM algorithms via PCA and Spectrum Clustering. Second, we propose an evaluation criterion using Linear Discriminative Analysis to choose the best local feature descriptions which plays a critical role in feature points detection. Third, we take advantage of multicore architecture of mobile terminal and parallelize the optimized SDM algorithm to improve the efficiency further. The experiment observations show that our final accomplished GPC‐SDM (improved Supervised Descent Method using spectrum clustering, PCA, and GPU acceleration) suppresses the memory usage, which is beneficial and efficient to meet the real‐time requirements.

An important step in the construction of a support vector machine (SVM) is to select optimal hyperparameters. This paper proposes a novel method for tuning the hyperparameters by maximizing the distance between two classes (DBTC) in the feature space. With a normalized kernel function, we find that DBTC can be used as a class separability criterion since the between-class separation and the within-class data distribution are implicitly taken into account. Employing DBTC as an objective function, we develop a gradient-based algorithm to search the optimal kernel parameter. On the basis of the geometric analysis and simulation results, we find that the optimal algorithm and the initialization problem become very simple. Experimental results on the synthetic and real-world data show that the proposed method consistently outperforms other existing hyperparameter tuning methods.