Variational Bayesian Gaussian Mixture Model Research Articles

Machine learning with electrocardiograms to optimize mortality risk stratification in patients with suspected acute coronary syndrome

Abstract Funding Acknowledgements Type of funding sources: Public grant(s) – National budget only. Main funding source(s): National Heart, Lung, and Blood Institute (NHLBI) and National Center for Advancing Translational Sciences (NCATS). Background The importance of risk stratification in patients with suspected acute coronary syndrome (ACS) extends beyond diagnosis and immediate treatment. It influences the precision of care delivery and the allocation of resources to those at highest risk of adverse events. Purpose We sought to evaluate the prognostic value of electrocardiogram (ECG) feature-based machine learning models to risk-stratify long-term mortality in those with suspected ACS. Methods This was a multicenter prospective observational cohort study of consecutive, non-traumatic patients evaluated at the emergency department for suspected ACS. The derivation cohort included 4,015 patients from a University Medical Center (age 59±16 years, 47% women, 80% training with 10-fold cross-validation and 20% testing). Ascertainment of all-cause death was based on numerous data sources, including the Centers for Disease Control and Prevention’s National Death Index registry. We trained six machine learning methods for survival analysis using 73 morphological ECG features from presenting prehospital 10-second 12-lead ECGs. We used predicted risk scores in a variational Bayesian Gaussian mixture model to define three risk groups corresponding to low, moderate, and high-risk and compared the resulting classification performance against the HEART score. The external testing cohort included 3,095 patients from a University (age 59±15 years, 44% women). Results The mortality rate was 20.3% in the derivation cohort after an average follow-up period of 3.53 years (IQR 1.75 - 5.32). Extra Survival Trees outperformed other forecasting models during cross-validation. In the internal testing cohort, the derived risk groups were significantly predictive of survival (Kaplan-Meier log-rank test statistic = 121.14, p<0.001), outperforming the classification performance using the HEART score (Figure 1). Our machine learning-based risk stratification model detected >90% of death events missed by low-risk HEART score (Figure 2). Ruling-out the low-risk group, it achieved a negative predictive value of 93.4% with a sensitivity of 85.9% (compared to 89.0% and 75.0%, respectively, for the HEART score). This classification performance generalized well to our external testing cohort. Compared to those in the low-risk group, patients at moderate (OR = 3.62 [1.35-9.74]) and high (OR = 6.12 [2.38-15.75]) risk had significantly higher odds of 30-day cardiovascular death (n=109, 3.5%). Conclusions Using only features from the 10-second 12-lead ECG, we derived and externally validated a machine learning model that stratifies the mortality risk of patients during long-term follow up. This model outperformed current standard of care based on the HEART score. Future work should validate the potential of this decision-support tool in influencing treatment plans and resource allocation.Figure 1.Clustering performanceFigure 2.Sankey diagram

Soft sensor development for mixed oil interface tracking in multi-product pipelines based on knowledge-informed semi-supervised Variational Bayesian Gaussian mixture regression

Sequential transportation of petroleum products in multi-product pipelines often lead to occurrence of mixed oil. The prediction of the arrival time of the mixed oil interface constitutes crucial data for the scheduling of treatment actions. However, existing soft sensors, like Gaussian mixture regression (GMR), may face challenges due to limited size of labeled data and numerical issues, leading to performance degradation or even training failure. To tackle these issues, we propose a Knowledge-informed Semi-supervised Variational Bayesian Gaussian mixture model (KI-SSVBGMR). It employs a semi-supervised fully Bayesian structure designed to address the constraints arising from the potential matrix singularity issues and shortage of labeled samples. Subsequently, we determine the crucial regression variable and establish its prior distribution based on industrial knowledge to improve model generalization. Finally, a learning procedure grounded in the Variational Inference algorithm is developed to train the KI-SSVBGMR. Through case studies, including numerical examples and real industrial datasets, our method demonstrates the effectiveness and reliability of the proposed soft sensor development method. This research can aid operators in improving mixed oil section management and provides valuable insights for integrating machine learning with industrial knowledge.

Abstract 1254: Integration of human omics analysis and new syngeneic tumor organoid models reveals that aberrant RAS/PI3K crosstalk triggers refractoriness in high-grade serous carcinoma

Abstract Refractory high-grade serous carcinoma (HGSC) is one of the most severe clinical problems in gynecology, but we have barely grasped the underlying mechanism of aggressiveness and chemo-resistance. Previous study has shown that homologous recombination genes, BRCA1/2-unmutated HGSC patients possess the potential to be in refractory condition, and their primary response to chemotherapy was heterogenous; thus, there can be subtype-specific determinant factors hatching refractoriness in BRCA1/2-unmutated HGSC patients. To explore the determinant factors, we performed gene set variation analysis and comprehensive log-rank tests on TCGA advanced serous ovarian cancer datasets. As a result, regulation of small GTPase signaling prominently correlated with poor prognosis uniquely in BRCA1/2-unmutated HGSC patients (p-value < 0.001), and eventually, pathway clustering analysis by variational Bayesian Gaussian mixture model discovered aberrant RAS/PI3K crosstalk as peculiar poor prognostic signature of BRCA1/2-unmutated HGSC patients. To elucidate the mechanism of poor progression triggered by RAS/PI3K, we utilized our syngeneic HGSC organoid models newly established from murine fallopian tube epithelium by making Myc overexpressing, and knocking out Trp53 and Rb1; as a supplement, the propriety of these genetic alterations to develop HGSC models was backed up by genetic profiles of high-grade serous ovarian cancer patients in AACR Project GENIE. After that, we knocked out Nf1 and Pten whose deletions were detected relatively with high frequency in HGSC to induce aberrant RAS/PI3K crosstalk in the primitive HGSC model. In-vivo evaluation and growth comparison indicated that aberrant RAS/PI3K crosstalk transformed Brca1/2-unmutated HGSC modeling cells into aggressive phenotypes such as poor survival (p-value < 0.01) and high growth rate (p-value < 0.001). Chemo-sensitivity assay also showed that the genetic manipulation made Brca1/2-unmutated HGSC modeling cells more tolerant to anti-tumor agents like carboplatin (p-value < 0.001), paclitaxel (p-value < 0.01), and olaparib (p-value < 0.01). To investigate biological processes that induce the aggressive and chemo-resistant phenotype, we performed gene set enrichment analysis among our HGSC models. The result showed that aberrant RAS/PI3K crosstalk induces cell cycle acceleration and downregulation of apoptosis in the HGSC cells, which backed up aggressive phenotype and chemo-resistance in our HGSC models. In conclusion, aberrant RAS/PI3K crosstalk transforms Brca1/2-unmutated HGSC modeling cells more aggressive and chemo-resistant, which was well recapitulated by our HGSC modeling organoids. The integration of our computational approach and experimental models will lead us to detect peculiar therapeutic targets against refractory HGSC. Citation Format: Tomohiro Tamura, Kenta Masuda, Shimpei Nagai, Keiyo Imaeda, Juntaro Yamasaki, Eiji Sugihara, Hiroyuki Nobusue, Yuji Otsuki, Rui Yamaguchi, Kazuhiro Sakurada, Wataru Yamagami, Hideyuki Saya, Osamu Nagano. Integration of human omics analysis and new syngeneic tumor organoid models reveals that aberrant RAS/PI3K crosstalk triggers refractoriness in high-grade serous carcinoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 1254.

Variational Bayesian Gaussian mixture model for off‐grid DOA estimation

AbstractWireless signals are commonly subject to diverse and complex noise interference. The typical assumption of Gaussian white noise often oversimplifies the noise, resulting in reduced accuracy in estimating the direction of arrival (DOA), especially in complex scenarios. To tackle this issue, this paper introduces a new Bayesian model for off‐grid DOA estimation. This model utilizes Gaussian mixture model (GMM)‐based Dirichlet processes (DP) to characterize noise, allowing adaptive adjustments in the number of Gaussian mixture models. Leveraging the factor graph representation of the Bayesian model, a low‐complexity mixed messaging passing algorithm, employing generalized approximate message passing (GAMP) and mean field (MF), is proposed. Simulation results validate the efficacy of the proposed algorithm.

Integrated Spatio-Temporal Deep Clustering (ISTDC) for cognitive workload assessment

Traditional high-dimensional electroencephalography (EEG) features (spectral or temporal) may not always attain satisfactory results in cognitive workload estimation. In contrast, deep representation learning (DRL) transforms high-dimensional data into cluster-friendly low-dimensional feature space. Therefore, this paper proposes an Integrated Spatio-Temporal Deep Clustering (ISTDC) model that uses DRL followed by a clustering method to achieve better clustering performance. The proposed model is illustrated using four Algorithms and Variational Bayesian Gaussian Mixture Model (VBGMM) clustering method. Temporal and spatial Variational Auto Encoder (VAE) models (mentioned in Algorithm 2 and Algorithm 3) learn temporal and spatial latent features from sequence-wise EEG signals and scalp topographical maps using the Long short-term memory and Convolutional Neural Network models. The concatenated spatio-temporal latent feature (mentioned in Algorithm 4) is passed to the VBGMM clustering method to efficiently estimate workload levels of n-back task. For the 0-back vs. 2-back task, the proposed model achieves the maximum mean clustering accuracy of 98.0%, and it improves by 11.0% over the state-of-the-art method. The results also indicate that the proposed multimodal approach outperforms temporal and spatial latent feature-based unimodal models in workload assessment.

Incremental Variational Bayesian Gaussian Mixture Model With Decremental Optimization for Distribution Accommodation and Fine-Scale Adaptive Process Monitoring.

Due to the frequent changes in operating conditions, time-varying behaviors, including slow-varying dynamics and switching modes, commonly exist in industrial processes, resulting in different degrees of shifting in the process data distribution. When the data distribution shifts in a relatively wide range, conventional adaptive methods become ineffective since they are unable to distinguish normal shifts from real faults, leading to false alarms. In this study, an incremental variational Bayesian Gaussian mixture model (IncVBGMM) is proposed for developing a fine-scale adaptive monitoring scheme to efficiently accommodate the shifting data distribution caused by different degrees of time-varying behaviors. First, IncVBGMM with decremental optimization is proposed to adapt to the changing data distribution via the automatic complement of local models while reducing redundancy to optimize the mixture model. Then, a fine-scale adaptive monitoring scheme is built with physical interpretations to discern between normal shifts and real faults by joint analysis of the static and dynamic information. In addition, a novel monitoring statistic called the expectation of variational Bayesian inference distance (EVBID) is proposed, which can quantify the distance from samples to the variational monitoring model and indicate the fault effects. Case studies involving a real-world three-phase flow facility reveal that the proposed method can accurately differentiate various types of faults from normal shifts and effectively adapt to the time-varying dynamics.

Two-stage data-driven dispatch for integrated power and natural gas systems by using stochastic model predictive control

The optimal dispatch of the integrated power and natural gas systems can increase the utilization rate of renewable energy and energy efficiency while decreasing operation costs. The common prediction errors of wind power and electric load have the potential to negatively impact the normal operation of the integrated power and natural gas systems. A two-stage data-driven dispatch strategy is proposed to reduce this effect, consisting of the day-ahead dispatch stage and the intraday rolling dispatch stage using stochastic model predictive control (MPC). In the day-ahead dispatch stage, the data-driven chance constraints of tie-line power and reserve of gas-fired generators are built, and the day-ahead tie-line power is obtained and regarded as input parameters to the intraday dispatch stage. In the intraday dispatch stage, the data-driven chance constraints of tie-line power and reserve of gas-fired generators with the latest rolling prediction data are built, and the remaining control variables are obtained. The distribution characteristics of the stochastic prediction errors of wind power and electric load are captured and described by the variational Bayesian Gaussian mixture model with massive historical data. Then the original stochastic mixed-integer nonlinear programming problem is converted to a tractable deterministic one by the quantile-based analytical reformulation and convex relaxation technique. Finally, the proposed strategy is verified by the numerical experiments based on a modified IEEE 33-bus system integrated with a 10-node natural gas system and a micro hydrogen system. The numerical results demonstrate that the proposed strategy reduces the actual costs and decreases the violation rate caused by the stochastic prediction errors of wind power and electric load.

Data‐driven chance‐constrained dispatch for integrated power and natural gas systems considering wind power prediction errors

AbstractStochastic wind power prediction errors hurt the normal operation of integrated power and natural gas systems (IPGS). First, the data‐driven stochastic chance‐constrained programming method is applied to deal with wind power prediction errors, and its probability distribution is accurately fitted by variational Bayesian Gaussian mixture model with massive historical data. In addition, the data‐driven chance constraints of tie‐line power and reserve capacity of gas turbine are built. Next, to utilize wind power more reasonably, the operational characteristics and optimal commitment of power‐to‐hydrogen devices are considered and modelled in proposed strategy to reflect the actual situation of IPGS. Then, the original complicated dispatch problem is converted into a tractable second‐order cone programming problem via convex relaxation and quantile‐based analytical reformulation techniques. Finally, the effectiveness of the proposed strategy is validated by numerical experiments based on a modified IEEE 33‐bus system integrated with a 10‐node natural gas system and a micro hydrogen system.

Heart failure with preserved ejection fraction phenogroup classification using machine learning.

Heart failure (HF) with preserved ejection fraction (HFpEF) is a complex syndrome with a poor prognosis. Phenotyping is required to identify subtype-dependent treatment strategies. Phenotypes of Japanese HFpEF patients are not fully elucidated, whose obesity is much less than Western patients. This study aimed to reveal model-based phenomapping using unsupervised machine learning (ML) for HFpEF in Japanese patients. We studied 365 patients with HFpEF (left ventricular ejection fraction >50%) as a derivation cohort from the Nara Registry and Analyses for Heart Failure (NARA-HF), which registered patients with hospitalization by acute decompensated HF. We used unsupervised ML with a variational Bayesian-Gaussian mixture model (VBGMM) with common clinical variables. We also performed hierarchical clustering on the derivation cohort. We adopted 230 patients in the Japanese Heart Failure Syndrome with Preserved Ejection Fraction Registry as the validation cohort for VBGMM. The primary endpoint was defined as all-cause death and HF readmission within 5years. Supervised ML was performed on the composite cohort of derivation and validation. The optimal number of clusters was three because of the probable distribution of VBGMM and the minimum Bayesian information criterion, and we stratified HFpEF into three phenogroups. Phenogroup 1 (n=125) was older (mean age 78.9±9.1years) and predominantly male (57.6%), with the worst kidney function (mean estimated glomerular filtration rate 28.5±9.7mL/min/1.73m2 ) and a high incidence of atherosclerotic factor. Phenogroup 2 (n=200) had older individuals (mean age 78.8±9.7years), the lowest body mass index (BMI; 22.78±3.94), and the highest incidence of women (57.5%) and atrial fibrillation (56.5%). Phenogroup 3 (n=40) was the youngest (mean age 63.5±11.2) and predominantly male (63.5±11.2), with the highest BMI (27.46±5.85) and a high incidence of left ventricular hypertrophy. We characterized these three phenogroups as atherosclerosis and chronic kidney disease, atrial fibrillation, and younger and left ventricular hypertrophy groups, respectively. At the primary endpoint, Phenogroup 1 demonstrated the worst prognosis (Phenogroups 1-3: 72.0% vs. 58.5% vs. 45%, P=0.0036). We also successfully classified a derivation cohort into three similar phenogroups using VBGMM. Hierarchical and supervised clustering successfully showed the reproducibility of the three phenogroups. ML could successfully stratify Japanese HFpEF patients into three phenogroups (atherosclerosis and chronic kidney disease, atrial fibrillation, and younger and left ventricular hypertrophy groups).

The load-haul-dump operation cycle recognition based on multi-sensor feature selection and bidirectional long short-term memory network

The operational cycle identification of the load-haul-dump (LHD) can help support the production process optimization in the underground mining industry and thus reduce mining costs. However, most of the existing research works use only the hydraulic bucket signal of LHD as the data source, and the stability and robustness of the identification method are poor. A few advanced research works use the variational Bayesian Gaussian mixture model to introduce other signals, but the accuracy of this recognition method is not perfect at present. In addition, the current identification methods are unable to simultaneously recognize the four working conditions of the LHD which include loading, hauling, dumping, and transiting. To solve these problems, a random forest feature selection (RFFS) and bidirectional long short-term memory (Bi-LSTM) based operation cycle recognition algorithm is proposed. Firstly, RFFS is used to remove redundant features based on the multi-sensor signals of the LHD. Then, Bi-LSTM is applied to fully exploit the temporal correlation between different operation regimes and accurately recognize the operation cycles. The effectiveness and superiority of the algorithm are verified by the experiment on the actual data of the LHD. The proposed algorithm can recognize four working conditions simultaneously, among which the recognition accuracy of loading conditions is the highest, up to 95.42%, and the weighted accuracy of this algorithm can reach 91.75% using the occupied time of each working condition as the weighting factor.

A machine learning-based detection framework against intermittent electricity theft attack

The widespread installation of advanced metering infrastructure (AMI) brings convenience to applications including but not limited to load management and demand response. However, AMI is also at risk of electricity theft and non-technical loss. Using the smart meter data provided by AMI to dig out user electricity consumption behavior is an effective way to construct electricity theft detectors. In this paper, a new intermittent electricity theft attack behavior is presented which switches between committing electricity theft and honestly consuming electricity alternately to skillfully evade the existing detectors. Based on the assumption that the labels of intermittent adversaries are unavailable, a new machine learning-based detection framework is proposed to detect this attack. Initially electricity features are constructed based on time intervals divided by a numerical iteration method. Then light gradient boosting method (LightGBM) is used to classify the normal users and adversaries. Further, the disperse degree of users is designed for capturing the differences between the intermittent adversaries and others. A new 2D label set is then constructed by combining the predicted labels of LightGBM and the disperse degree. Finally, a variational Bayesian Gaussian mixture model is employed based on the 2D label set to sort the users into the normal users and adversaries visually. Results of the case studies show that the presented attack can evade state-of-the-art detectors but still gain high profits. In addition, the proposed machine learning-based detector outperforms state-of-the-art detectors on both persistent attacks and intermittent attacks.

Stacking regression technology with event profile for electric vehicle fast charging behavior prediction

Large-scale deployment of electric vehicles (EVs) poses a huge challenge to the operation of the distribution network. As a possible mobile energy carrier, the interaction between EVs and distribution networks can provide some opportunities for power operation. Where to charge and how to charge have become an important research topic in EV charging scheduling. Previous studies mainly focused on slow-charging behavior analysis rather than fast-charging behavior. Here, we provide an in-depth understanding of EV user fast-charging behavior in public stations based on more than 220,000 real-world charging records with the Variational-Bayesian Gaussian-mixture model. Characteristics related to charging energy and charging duration are mainly considered in the cluster model, especially dwelling duration after charging is taken into account to better support the decision of charging recommendation strategy and charging power allocation. Inspired by the future application scenario of the charging behavior cluster of previous studies, we propose a charging behavior prediction framework considering behavior catalogues with stacking regression technology. The results show that the proposed framework improves the prediction accuracy of charging behavior and can effectively evaluate the priority of charging behavior.

Energy consumption mode identification and monitoring method of process industry system under unstable working conditions

Process industry systems under unstable working conditions are prone to potential anomalies, deviating from the original transition trajectory, and taking longer than expected to return to stability due to persistent disturbances from uncertainties and experience-based regulation errors. The energy waste caused by this situation has not received sufficient attention, and cannot be addressed by existing energy consumption monitoring methods. Herein, an energy consumption mode (ECM) identification and monitoring method under unstable working conditions is proposed, consisting of ECM identification model and multi-mode dynamic monitoring model, focusing on the variation rules of the correlation between energy consumption and other states of the system. In the ECM identification stage, the ECM correlation parameters that reflect the comprehensive production information are selected. Then, given the transfer characteristics of ECM, a Hidden Semi-Markov Model (HSMM) is constructed to fit the migration between modes and the duration within modes. The Variational Bayesian Gaussian Mixture Model is introduced to improve the HSMM, which solves the problem of lacking prior knowledge of ECM and achieves the automatic classification and online identification of ECM. In the dynamic monitoring stage of multi-ECMs, a series of dynamic kernel principle component analysis models are established, and the corresponding monitoring thresholds are set for each ECM. By calculating the maximum of the posteriori probability and the mode thresholds, the ECMs under unstable conditions can be accurately identified and automatically monitored. Compared with previous methods, the proposed method reduces the false detection rate and missed detection rate of abnormal ECM identification to 1.04% and 1.31% in the actual slag grinding production process, which proves its effectiveness.

HT-Fed-GAN: Federated Generative Model for Decentralized Tabular Data Synthesis.

In this paper, we study the problem of privacy-preserving data synthesis (PPDS) for tabular data in a distributed multi-party environment. In a decentralized setting, for PPDS, federated generative models with differential privacy are used by the existing methods. Unfortunately, the existing models apply only to images or text data and not to tabular data. Unlike images, tabular data usually consist of mixed data types (discrete and continuous attributes) and real-world datasets with highly imbalanced data distributions. Existing methods hardly model such scenarios due to the multimodal distributions in the decentralized continuous columns and highly imbalanced categorical attributes of the clients. To solve these problems, we propose a federated generative model for decentralized tabular data synthesis (HT-Fed-GAN). There are three important parts of HT-Fed-GAN: the federated variational Bayesian Gaussian mixture model (Fed-VB-GMM), which is designed to solve the problem of multimodal distributions; federated conditional one-hot encoding with conditional sampling for global categorical attribute representation and rebalancing; and a privacy consumption-based federated conditional GAN for privacy-preserving decentralized data modeling. The experimental results on five real-world datasets show that HT-Fed-GAN obtains the best trade-off between the data utility and privacy level. For the data utility, the tables generated by HT-Fed-GAN are the most statistically similar to the original tables and the evaluation scores show that HT-Fed-GAN outperforms the state-of-the-art model in terms of machine learning tasks.

A Variational Bayesian Gaussian Mixture-Nonnegative Matrix Factorization Model to Extract Movement Primitives for Robust Control

Nonnegative matrix factorization (NMF) is a powerful tool for parameter estimation applied in numerous robotics applications, such as path planning, motion trajectory prediction, and motion intention detection. In particular, NMF has been successfully used to extract simplified and organized movement primitives from myoelectric signal (MES) for robust control of multi-degree of freedom humanoid robots. However, MES is typically contaminated by complex noise sources. The system performance often degrades due to the simplified Gaussian assumption of the noise distribution in existing NMF methods. Furthermore, most existing NMF models are unable to automatically determine the rank of the latent matrices. To address these issues, this article presents a hybrid variational Bayesian Gaussian mixture and NMF (GMNMF) model with a finite Gaussian mixture model adopted to fit the mixed noise density function of MES. In addition, the automatic relevant determination criterion is applied to automatically infer the number of movement primitives. The coordinate descent update rules for the proposed model are formulated by mean-field variational Bayesian inference. We assess the model performance on five synthetic noise distribution functions and an experimental MES dataset to perform six wrist movements. The results demonstrate that GMNMF yields low error and high robustness in extracting the movement primitives over four competitive methods for robust cybernetic control.

A variational Bayesian approach to multiantenna spectrum sensing under correlated noise

In this paper, we address the problem of spectrum sensing when the secondary user (SU) is equipped with a multiantenna receiver and noise across antenna elements are correlated. Covariance matrix of correlated noise is non-diagonal and its standard-condition-number (SCN) is much greater than one, which results in significantly degraded detection performance. Many of the existing techniques either whiten the covariance matrix or utilize geometrical properties of noise covariance matrix for the design of decision statistic. However, these techniques have low detection performance in uncorrelated noise or when the correlation is low. To address this problem and further improve the detection performance, a novel spectrum sensing algorithm based on variational Bayesian Gaussian mixture model (VBGMM) is proposed. Specifically, a new decision vector is proposed which is amenable to VBGMM and incorporates signal features suitable for spectrum sensing in both correlated and uncorrelated noise. Decision vectors extracted from the received signal are clustered using variational expectation-maximization (VEM) algorithm to detect the presence of primary user (PU). Compared to existing techniques, the proposed method is shown by simulation to achieve higher detection rates for the entire range of correlation.

Understanding Urban Area Attractiveness Based on Private Car Trajectory Data Using a Deep Learning Approach

With the fast development of urbanization and motorization, an increasing number of people choose to buy private cars to fulfill their daily travel needs. In particular, many people from various positions of the city drive their cars to specified areas, and then they will stop and stay for a certain period of time, leading to a spatiotemporal evolution of urban area attractiveness (AA). In this paper, we aim at understanding urban AA based on analyses of private car trajectory datasets. Specifically, by extracting point-of-stop (PoS) data from the private car trajectories, we design the variational Bayesian Gaussian mixture models (VBGMM) to deduce the probability density distribution of PoSs and connect it to the variation of AA. We establish a deep learning model based on long short-term memory (LSTM) to capture the evolution of the AA. Furthermore, we integrate dropout in the LSTM method to address challenging issues such as overfitting and time-consuming training of complex neural networks in the AA prediction. We conduct experiments by using real-world private car trajectory data to evaluate the performance of the proposed method. The results validate that our proposed method outperforms existing ones in terms of various metrics. To the authors’ knowledge, our work is the first one to utilize private car trajectory data to study urban area attractiveness, thereby facilitating a new perspective regarding an understanding of human travel behavior and the evolution of urban mobility.

Machine Learning-Based Multipath Components Clustering and Cluster Characteristics Analysis in High-Speed Railway Scenarios

Channel multipath components (MPCs) clustering and cluster characterization are the prerequisite for the development of cluster based channel models. This article investigates the MPCs clustering based on machine learning (ML) and analyzes the cluster characteristics in typical high-speed railway (HSR) scenarios. A variational Bayesian Gaussian mixture model (GMM)-based algorithm is introduced to achieve the space–time clustering of MPCs, which incorporates the statistical characteristics of MPCs and can automatically determine the optimum number of clusters. Moreover, a novel density-based validity index is proposed for evaluating the MPCs clustering performance. The proposed validity index improves the traditional index by considering the intracluster density, which can be calculated according to the Graham scanning method and Green’s formula. In addition to synthetic datasets, realistic MPCs datasets collected in an HSR obstructed viaduct scenario are used for the performance evaluation of the clustering algorithm and clustering validity index. Based on clustering results in the measured scenario, static cluster characteristics, including cluster number, intercluster and intracluster delay spreads (DSs) and angle spreads (ASs), and dynamic cluster characteristics such as cluster lifetime and birth and death property of clusters, are extracted and analyzed. These results will be useful for cluster-based channel modeling in future HSR mobile communication systems.

Identifying evolving leading edge erosion by tracking clusters of lift coefficients

This work proposes an approach to identify Leading Edge Erosion (LEE) of a wind turbine blade by tracking evolving and emerging clusters of lift coefficients CL time-series signals under uncertain inflow conditions. Most diagnostic techniques today rely on direct visual inspection, image processing, and statistical analysis, e.g. data mining or regression on SCADA output signals. We claim that probabilistic multivariate spatio-temporal techniques could play an eminent role in the diagnostics of LEE specifically leveraging CL time-series signals form multiple sections along the span of the blade. The proposed method extracts clusters’ features based on Variational Bayesian Gaussian Mixture Models (VBGMM) and tracks their spatial and temporal changes, as well as interpret the evolution of the clusters through prior physics-based assumptions. The parameters of the VBGMM are the mean, the eigenvalues and eigenvectors of the covariance matrix, and the angle of orientation of the eigenvectors. We show that the distribution of the CL data may not show statistically separable clusters, however, the parameters of the VBGMM clusters fitted to the CL data, allows to discriminate moving clusters primarily due to varying inflow and operating conditions, versus emerging clusters primarily due to evolving severity of the blade LEE.

DRKPCA-VBGMM: fault monitoring via dynamically-recursive kernel principal component analysis with variational Bayesian Gaussian mixture model

DRKPCA-VBGMM: fault monitoring via dynamically-recursive kernel principal component analysis with variational Bayesian Gaussian mixture model