Practical Datasets Research Articles

FedSTN: Graph Representation Driven Federated Learning for Edge Computing Enabled Urban Traffic Flow Prediction

Predicting traffic flow plays an important role in reducing traffic congestion and improving transportation efficiency for smart cities. Traffic Flow Prediction (TFP) in the smart city requires efficient models, highly reliable networks, and data privacy. As traffic data, traffic trajectory can be transformed into a graph representation, so as to mine the spatio-temporal information of the graph for TFP. However, most existing work adopt a central training mode where the privacy problem brought by the distributed traffic data is not considered. In this paper, we propose a Federated Deep Learning based on the Spatial-Temporal Long and Short-Term Networks (FedSTN) algorithm to predict traffic flow by utilizing observed historical traffic data. In FedSTN, each local TFP model deployed in an edge computing server includes three main components, namely Recurrent Long-term Capture Network (RLCN) module, Attentive Mechanism Federated Network (AMFN) module, and Semantic Capture Network (SCN) module. RLCN can capture the long-term spatial-temporal information in each area. AMFN shares short-term spatio-temporal hidden information when it trains its local TFP model by the additive homomorphic encryption approach based on Vertical Federated Learning (VFL). We employ SCN to capture semantic features such as irregular non-Euclidean connections and Point of Interest (POI). Compared with existing baselines, several simulations are conducted on practical data sets and the results prove the effectiveness of our algorithm.

Spatial Analysis of the PM2.5 Air Pollution in Iraq by using R Language

Abstract Remote sensing refers to the acquisition of data and information about a phenomenon and a territory, without a direct contact with it. It is alternative to in-situ observation. Due to its wide coverage, multi-temporal and multispectral capability, remote sensing satellites have been used for studying and monitoring the spatial distribution of the aerosol pollution (like PM2.5) in the atmosphere. PM2.5 refers to atmospheric particulate matter (PM) that has a diameter of less than 2.5 micrometers, which is about 3% the diameter of a human hair. The current study concentrates on the using of R programming language to process and analyze the retrieval PM2.5 dataset over Iraq for the period (January 1, 2020 to December 31, 2020), these data were downloaded from the European Union’s Earth observation Copernicus program through the Copernicus Atmosphere Monitoring Service (CAMs). The study presented output results of the PM2.5 concentration levels by developing an automatic tool that restructuring the netCDF file into excel file using R language. In addition, it produces different valuable charts that help in practical dataset interpretation and decision-making. The results of the study showed the variation in the spatial distribution of PM2.5 in the study area due to the diversity of pollution sources from one region to another.

Digital twin for intelligent probabilistic short term load forecasting in solar based smart grids using shark algorithm

This article proposes a novel evolving based prediction model for the accurate short term load forecasting in solar based smart grids. The proposed method uses a probabilistic method for uncertainty quantization to make sure that the maximum modeling of the prediction interval would be achieved in a renewable based environment. In this regard, the innovative lower upper bound estimation method (LUBE) is trained using the real-time data of the smart grid gathered by the digital twin of the system. This would result in much higher results due to the avoidance of malfunction of the smart metering devices located within the smart grid. Digital twin can help predict the load demand in solar based smart grids by using machine learning algorithms to analyze the data from the smart grid. This data can be used to create a model of the system and predict how the load demand will change based on different factors such as weather, time of day, and season. By understanding the load demand, solar based smart grids can better manage their energy resources and optimize the performance of the system. In order to improve the model performance, white shark optimization algorithm (WSOA) is used as the trainer of the prediction model in a heuristic environment. The results advocate the high accuracy and reliability of the proposed method on practical dataset,

First-order spatial dependent count integer-valued autoregressive (Sp-DCINAR(1,1)) process

In this article, we propose a new model to model the spatial count data on a two-dimensional regular grid using binomial thinning operator with dependent Bernoulli counting series. The model is called “first-order spatial dependent count integer-valued autoregressive (Sp-DCINAR(1,1)) model.” Some of its properties have been derived and the estimation of the parameters of the model are obtained by the Yule-Walker method, conditional least squares method and conditional maximum likelihood method. Finally, numerical results are presented together with an application to a two-dimensional practical data set.

Navigation pattern extraction from AIS trajectory big data via topic model

AbstractThis paper introduces a novel approach for extracting vessel navigation patterns from very large automatic identification system (AIS) trajectory big data. AIS trajectory data records are first converted to a series of code documents using vector quantisation, such as k-means and PQk-means algorithms, whose performance is evaluated in terms of precision and computational time. Therefore, a topic model is applied to these code documents from which vessels’ navigation patterns are extracted and identified. The potential of the proposed approach is illustrated by several experiments conducted with a practical AIS dataset in a region of North West France. These experimental results show that the proposed approach is highly appropriate for mining AIS trajectory big data and outperforms common DBSCAN algorithms and Gaussian mixture models.

Contactless Sensing-Aided Respiration Signal Acquisition Using Improved Empirical Wavelet Transform for Rhythm Detection.

Respiration is one of the most important vital signs indicating physical condition, while the signal detection is challenging due to the complex rhythm and effort in practical scenarios. In this paper, we propose a contactless sensing-aided respiration signal acquisition technique, which can adaptively extract the desired signal under time-varying respiration rhythms within a wide range. To be specific, respiration is perceived by piezoelectric ceramics sensors along with ballistocardiography and other interference in a contactless manner, and the proposed improved empirical wavelet transform (IEWT) performs spectrum division and recognition based on upper envelop and principal component criteria, respectively, to adaptively extract the respiration spectrum for signal reconstruction. For validations, we extracted respiration signals from 8 healthy individuals in lab breathing at specified rhythms from 0.2 Hz to 0.6 Hz as well as 38 in-patients suffering from sleep-disordered-breathing with reference of polysomnogram in practical clinic scenario. The results showed that the detected respiration rhythms perfectly fitted the ones in experimental lab dataset with a correlation coefficient of 0.98, which validated the effectiveness of the respiration spectrum extraction of the proposed IEWT method. Besides, in practical clinical dataset, the proposed IEWT method could yield mean absolute and relative errors of respiration intervals of 0.4 and 0.05 seconds, respectively, achieving significant improvement in comparison with conventional ones. Meanwhile, the performance of IEWT was robust to rhythm variation, individual difference and breathing cycle detection techniques, which demonstrated the feasibility and superiority of the proposed IEWT method for practical respiration monitoring.

A Dataset Auditing Method for Collaboratively Trained Machine Learning Models.

Dataset auditing for machine learning (ML) models is a method to evaluate if a given dataset is used in training a model. In a Federated Learning setting where multiple institutions collaboratively train a model with their decentralized private datasets, dataset auditing can facilitate the enforcement of regulations, which provide rules for preserving privacy, but also allow users to revoke authorizations and remove their data from collaboratively trained models. This paper first proposes a set of requirements for a practical dataset auditing method, and then present a novel dataset auditing method called Ensembled Membership Auditing (EMA). Its key idea is to leverage previously proposed Membership Inference Attack methods and to aggregate data-wise membership scores using statistic testing to audit a dataset for a ML model. We have experimentally evaluated the proposed approach with benchmark datasets, as well as 4 X-ray datasets (CBIS-DDSM, COVIDx, Child-XRay, and CXR-NIH) and 3 dermatology datasets (DERM7pt, HAM10000, and PAD-UFES-20). Our results show that EMA meet the requirements substantially better than the previous state-of-the-art method. Our code is at: https://github.com/Hazelsuko07/EMA.

A novel domain generalization network with multidomain specific auxiliary classifiers for machinery fault diagnosis under unseen working conditions

The domain adaptation-based intelligent diagnosis approaches have achieved promising performance on diagnosis tasks under different working conditions. However, these methods rely on a premise that the target data are available in the model training phase. In real industries, collecting interest data from target machines in advance may be infeasible, which greatly restricts the practicality of intelligent diagnosis approaches in reality. To solve this issue, this study proposes a novel domain generalization network for machinery fault diagnosis where interest data are completely unavailable during model training. In the proposed network, multiple domain-specific auxiliary classifiers are firstly designed to effectively learn domain-specific features from each source domain, and then, a convolutional auto-encoder module is further constructed to map raw signals into a new feature space where the learned domain-specific features are removed. Meanwhile, with the features outputted by the convolutional auto-encoder, a domain-invariant classifier with inter-domain alignment strategy is designed to learn generalization diagnostic knowledge among different source domains, thereby performing diagnosis tasks under unseen conditions. Experiments on three practical rotary machinery datasets validate the effectiveness of the proposed network, showing that the proposed network is promising for fault diagnosis tasks in practical scenarios.

Spatio-Temporal Self-Supervised Learning for Traffic Flow Prediction

Robust prediction of citywide traffic flows at different time periods plays a crucial role in intelligent transportation systems. While previous work has made great efforts to model spatio-temporal correlations, existing methods still suffer from two key limitations: i) Most models collectively predict all regions' flows without accounting for spatial heterogeneity, i.e., different regions may have skewed traffic flow distributions. ii) These models fail to capture the temporal heterogeneity induced by time-varying traffic patterns, as they typically model temporal correlations with a shared parameterized space for all time periods. To tackle these challenges, we propose a novel Spatio-Temporal Self-Supervised Learning (ST-SSL) traffic prediction framework which enhances the traffic pattern representations to be reflective of both spatial and temporal heterogeneity, with auxiliary self-supervised learning paradigms. Specifically, our ST-SSL is built over an integrated module with temporal and spatial convolutions for encoding the information across space and time. To achieve the adaptive spatio-temporal self-supervised learning, our ST-SSL first performs the adaptive augmentation over the traffic flow graph data at both attribute- and structure-levels. On top of the augmented traffic graph, two SSL auxiliary tasks are constructed to supplement the main traffic prediction task with spatial and temporal heterogeneity-aware augmentation. Experiments on four benchmark datasets demonstrate that ST-SSL consistently outperforms various state-of-the-art baselines. Since spatio-temporal heterogeneity widely exists in practical datasets, the proposed framework may also cast light on other spatial-temporal applications. Model implementation is available at https://github.com/Echo-Ji/ST-SSL.

A dynamic predictive maintenance approach using probabilistic deep learning for a fleet of multi-component systems

Empowered by the ubiquitous sensing infrastructure, predictive maintenance (PdM) has received increasing attention. However, designing predictive maintenance for multiple systems remains a challenging problem as it requires effective integration of the prediction and scheduling under uncertainty and constraints. This paper proposes a dynamic PdM approach for a fleet of multi-component systems, which integrates deep learning-based (DL-based) probabilistic remaining useful life (RUL) prognostics into maintenance planning. First, a DL-based framework is designed for obtaining predictive distributions of RULs, and a probabilistic gated recurrent unit model is proposed for time series data. The Gaussian distribution is used to model prediction uncertainties, and the negative log-likelihood loss is used for model training, which inclines to emphasize the late degradation period. Then, an integer program is established and embedded in a rolling horizon approach for maintenance optimization. The concept of wasted residual values is proposed to quantify the impact of early retirement of components on total costs, which enables customized group maintenance planning under different cost settings. The applicability of the overall approach is illustrated for a fleet of aircraft based on a practical dataset. The experimental results show that our approach can provide accurate RUL prognostics and schedule maintenance flexibly to reduce costs.

Optimal Scheduling of Rainwater Collection Vehicles: Mixed Integer Programming and Genetic Algorithms

Due to climate change, some areas in the world witnessed higher levels of heavy rain than the capacity of the wastewater system of the streets. Therefore, water tankers are used for the dewatering process to take the extra rainwater from the streets to keep a smooth flow of vehicles and to use the water in agriculture and industry. Water is taken to a water treatment plant. Performing the dewatering process as fast as possible, especially in crowded streets, was ignored by researchers. In this study, at first, the problem was solved using two mixed integer programming (MIP) models. A new variant of identical parallel machine scheduling with job splitting is proposed for the first time, where one or at most two tankers can work at the same flood location at the same time. This is performed in the second model. However, the first model considers dividing the dewatering processes into two phases, where the first one, which is more urgent, is to reduce the amount of floodwater. The second one is for dewatering the rest of the water. Then two genetic algorithms (GAs) were used to solve faster the two MIP models, which are NP-hard problems. At first, the MIP and GA models were applied to small-sized problems. Then GA was used for large practical data sets. Results showed that for small problems, MIP and GA gave optimal solutions in a reasonable number of iterations, while for larger problems, good solutions were obtained in a reasonable number of iterations.

MrCAN: Multi-relations aware convolutional attention network for multivariate time series forecasting

Multivariate time series forecasting (MTSF) has gathered extensive attention in various research areas. Many researchers leverage deep neural networks to explore spatial–temporal relationships for MTSF with great success. Nevertheless, plentifully available data required by deep neural networks often struggle to satisfy practical scenarios. To overcome this limitation, we exploit the internal structure of deep neural networks to automatically learn sample relationships, aiming to mitigate data scarcity by propagating information among different data samples. Consequently, in this paper, we propose a Multi-Relations aware Convolutional Attention Network, termed MrCAN, which integrates spatial–temporal relation learning and sample relation learning, ad hoc for addressing MTSF issues. In MrCAN, we propose a novel spatial–temporal attention module for spatial–temporal relationship representation learning. With the proposed batch attention module, we further explore relational modeling among samples in each mini-batch to implicitly enrich the training sample information. In particular, we introduce parameter-sharing regressors located before and after the batch attention module to alleviate the training and testing inconsistency in learning batch invariant representations. Extensive experiments on six practical datasets demonstrate that MrCAN compares favorably to nine baseline models. Larger performance gaps are exhibited, especially on small datasets like the SML2010 dataset. Code is publicly available at https://github.com/JZhangNA/MrCAN.

Maritime Target Saliency Detection for UAV Based on the Stimulation Competition Selection Mechanism of Raptor Vision

A maritime target saliency detection method inspired by the stimulation competition and selection mechanism of raptor vision is presented for the airborne vision system of unmanned aerial vehicle (UAV) in an unknown maritime environment. The stimulation competition and selection mechanism in the visual pathway of raptor vision based on the phenomenon of raptor capturing prey in complex scenes are studied. Then, the mathematical model of the stimulation competition and selection mechanism of raptor vision is established and employed for the salient object detection. Popular image datasets and practical scene datasets are applied to verify the effectiveness of the presented method. Results show that the detection performance of the proposed method is better than that of other comparison methods. The proposed algorithm provides an idea for maritime target salient detection and cross-domain joint mission for UAV or other unmanned equipment.

Noise Suppression With Similarity-Based Self-Supervised Deep Learning.

Image denoising is a prerequisite for downstream tasks in many fields. Low-dose and photon-counting computed tomography (CT) denoising can optimize diagnostic performance at minimized radiation dose. Supervised deep denoising methods are popular but require paired clean or noisy samples that are often unavailable in practice. Limited by the independent noise assumption, current self-supervised denoising methods cannot process correlated noises as in CT images. Here we propose the first-of-its-kind similarity-based self-supervised deep denoising approach, referred to as Noise2Sim, that works in a nonlocal and nonlinear fashion to suppress not only independent but also correlated noises. Theoretically, Noise2Sim is asymptotically equivalent to supervised learning methods under mild conditions. Experimentally, Nosie2Sim recovers intrinsic features from noisy low-dose CT and photon-counting CT images as effectively as or even better than supervised learning methods on practical datasets visually, quantitatively and statistically. Noise2Sim is a general self-supervised denoising approach and has great potential in diverse applications.

BCE-Net: Reliable building footprints change extraction based on historical map and up-to-date images using contrastive learning

Automatic and periodic recompiling of building databases with up-to-date high-resolution images has become a critical requirement for rapidly developing urban environments. However, the architecture of most existing approaches for change extraction attempts to learn features related to changes but ignores objectives related to buildings. This inevitably leads to the generation of significant pseudo-changes, due to factors such as seasonal changes in images and the inclination of building façades. To alleviate the above-mentioned problems, we developed a contrastive learning approach by validating historical building footprints against single up-to-date remotely sensed images. This contrastive learning strategy allowed us to inject the semantics of buildings into a pipeline for the detection of changes, which is achieved by increasing the distinguishability of features of buildings from those of non-buildings. In addition, to reduce the effects of inconsistencies between historical building polygons and buildings in up-to-date images, we employed a deformable convolutional neural network to learn offsets intuitively. In summary, we formulated a multi-branch building extraction method that identifies newly constructed and removed buildings, respectively. To validate our method, we conducted comparative experiments using the public Wuhan University building change detection dataset and a more practical dataset named SI-BU that we established. Our method achieved F1 scores of 93.99% and 70.74% on the above datasets, respectively. Moreover, when the data of the public dataset were divided in the same manner as in previous related studies, our method achieved an F1 score of 94.63%, which surpasses that of the state-of-the-art method. Code and datasets are available at https://vrlab.org.cn/~hanhu/projects/bcenet.

Zero-inflated probit Bernoulli model: a new model for binary data

This study proposes a new model in the family zero-inflated (ZI) binary models, it is called as a ZI Probit Bernoulli (ZIPBer) model. This model can be performed to model or simulate binary data that has an excessive amount of zero counts. At the framework of this research, we first introduce the classical formula, estimating equation (EE) and related functions for the ZIPBer model. Next, the results of asymptotic inferences for the ZIPBer model is studied based on the widespread regularity conditions. For the numerical illustrations, many simulation studies and a practical data set are executed in this paper. Specifically, we perform the fishing data set (FDS) to check the working effectiveness and robustness of the maximum likelihood estimation (MLE) approach in estimating parameters for the ZIPBer model. We also illustrate that this actual data set is more suitable for the proposed model: ZIPBer than other binary models such as logit, probit and ZIBer. The achieved results in this study are very unanimous with reality. It also illustrates that this is a dependable verification for us to know the solution as well as how to proceed to how to catch or take the most fish. This is the contribution of this work in terms of application. Finally, some crucial discussions and conclusions are given in this paper.

An Intelligent Boosting and Decision-Tree-Regression-Based Score Prediction (BDTR-SP) Method in the Reform of Tertiary Education Teaching

The reform of tertiary education teaching promotes teachers to adjust timely teaching plans based on students’ learning feedback in order to improve teaching performance. Thefore, learning score prediction is a key issue in process of the reform of tertiary education teaching. With the development of information and management technologies, a lot of teaching data are generated as the scale of online and offline education expands. However, a teacher or educator does not have a comprehensive dataset in practice, which challenges his/her ability to predict the students’ learning performance from the individual’s viewpoint. How to overcome the drawbacks of small samples is an open issue. To this end, it is desirable that an effective artificial intelligent tool is designed to help teachers or educators predict students’ scores well. We propose a boosting and decision-tree-regression-based score prediction (BDTR-SP) model, which relies on an ensemble learning structure with base learners of decision tree regression (DTR) to improve the prediction accuracy. Experiments on small samples are conducted to examine the important features that affect students’ scores. The results show that the proposed model has advantages over its peer in terms of prediction correctness. Moreover, the predicted results are consistent with the actual facts implied in the original dataset. The proposed BDTR-SP method aids teachers and students to predict students’ performance in the on-going courses in order to adjust the teaching and learning strategies, plans and practices in advance, enhancing the teaching and learning quality. Therefore, the integration of information technology and artificial intelligence into teaching and learning practices is able to push forward the reform of tertiary education teaching.

Weakly supervised real-time instance segmentation for ultrasound images of median nerves

Medical ultrasound technology has garnered significant attention in recent years, with Ultrasound-guided regional anesthesia (UGRA) and carpal tunnel diagnosis (CTS) being two notable examples. Instance segmentation, based on deep learning approaches, is a promising choice to support the analysis of ultrasound data. However, many instance segmentation models cannot achieve the requirement of ultrasound technology e.g. real-time. Moreover, fully supervised instance segmentation models require large numbers of images and corresponding mask annotations for training, which can be time-consuming and labor-intensive in the case of medical ultrasound data. This paper proposes a novel weakly supervised framework, CoarseInst, to achieve real-time instance segmentation of ultrasound images with only box annotations. CoarseInst not only improves the network structure, but also proposes a two-stage “coarse-to-fine” training strategy. Specifically, median nerves are used as the target application for UGRA and CTS. CoarseInst consists of two stages, with pseudo mask labels generated in the coarse mask generation stage for self-training. An object enhancement block is incorporated to mitigate the performance loss caused by parameter reduction in this stage. Additionally, we introduce a pair of loss functions, the amplification loss, and the deflation loss, that work together to generate the masks. A center area mask searching algorithm is also proposed to generate labels for the deflation loss. In the self-training stage, a novel self-feature similarity loss is designed to generate more precise masks. Experimental results on a practical ultrasound dataset demonstrate that CoarseInst could achieve better performance than some state-of-the-art fully supervised works.

A New Probability Distribution: Model, Theory and Analyzing the Recovery Time Data

Probability models are frequently used in numerous healthcare, sports, and policy studies. These probability models use datasets to identify patterns, analyze lifetime scenarios, predict outcomes of interest, etc. Therefore, numerous probability models have been studied, introduced, and implemented. In this paper, we also propose a novel probability model for analyzing data in different sectors, particularly in biomedical and sports sciences. The probability model is called a new modified exponential-Weibull distribution. The heavy-tailed characteristics along with some other mathematical properties are derived. Furthermore, the estimators of the new modified exponential-Weibull are derived. A simulation study of the new modified exponential-Weibull model is also provided. To illustrate the new modified exponential-Weibull model, a practical dataset is analyzed. The dataset consists of seventy-eight observations and represents the recovery time after the injuries in different basketball matches.

Identifying High-Need Primary Care Patients Using Nursing Knowledge and Machine Learning Methods.

Patient cohorts generated by machine learning can be enhanced with clinical knowledge to increase translational value and provide a practical approach to patient segmentation based on a mix of medical, behavioral, and social factors. This study aimed to generate a pragmatic example of how machine learning could be used to quickly and meaningfully cohort patients using unsupervised classification methods. Additionally, to demonstrate increased translational value of machine learning models through the integration of nursing knowledge. A primary care practice dataset (N = 3,438) of high-need patients defined by practice criteria was parsed to a subset population of patients with diabetes (n = 1233). Three expert nurses selected variables for k-means cluster analysis using knowledge of critical factors for care coordination. Nursing knowledge was again applied to describe the psychosocial phenotypes in four prominent clusters, aligned with social and medical care plans. Four distinct clusters interpreted and mapped to psychosocial need profiles, allowing for immediate translation to clinical practice through the creation of actionable social and medical care plans. (1) A large cluster of racially diverse female, non-English speakers with low medical complexity, and history of childhood illness; (2) a large cluster of English speakers with significant comorbidities (obesity and respiratory disease); (3) a small cluster of males with substance use disorder and significant comorbidities (mental health, liver and cardiovascular disease) who frequently visit the hospital; and (4) a moderate cluster of older, racially diverse patients with renal failure. This manuscript provides a practical method for analysis of primary care practice data using machine learning in tandem with expert clinical knowledge.