Combining Dimension Reduction Research Articles

Exploring Attribute Associations in Pedestrian-Involved Hit-and-Run Crashes through Cluster Correspondence Analysis

Pedestrian-involved hit-and-run (PIHR) crashes represent a significant public health concern, and identifying patterns in these crashes can aid in developing effective countermeasures. Cluster correspondence analysis (CCA) is a multidimensional statistical technique that combines dimension reduction and clustering to identify patterns in categorical data. This method provides insights into underlying patterns and relationships among categories. The current study analyzed a Louisiana crash dataset of 2,201 PIHR crashes from 2015 to 2019 using CCA to identify underlying patterns. CCA identified six clusters, examined the top associative attributes, and assessed their cluster-to-dataset percentage ratio. The first two clusters, representing 66% of PIHR crashes, mainly involved crashes on city streets, occurring primarily during early night (7 to 11 p.m.) in Cluster 1 and the afternoon (12 noon to 4 p.m.) in Cluster 2. Clusters 3 and 4, accounting for 30% of PIHR crashes, predominantly exhibited crashes on U.S. and state highways. Cluster 4, which featured fatalities, primarily concentrated on state highways during the early morning hours (4 to 6 a.m.). Meanwhile, Clusters 5 and 6 focused on high-speed highways, specifically interstates involving pedestrian fatalities. A discussion on implementing strategic countermeasures tailored to the distinct characteristics of each cluster is presented. Alongside improvements in context-based countermeasures to ease pedestrian movement and enhance their visibility, strategies such as advocating for stringent hit-and-run laws, incentivizing the use of dashcams, and broadly publicizing resources for crash reporting are projected to be highly effective in curbing PIHR crashes.

A comparative study of feature selection and feature extraction methods for financial distress identification

Financial distress identification remains an essential topic in the scientific literature due to its importance for society and the economy. The advancements in information technology and the escalating volume of stored data have led to the emergence of financial distress that transcends the realm of financial statements and its’ indicators (ratios). The feature space could be expanded by incorporating new perspectives on feature data categories such as macroeconomics, sectors, social, board, management, judicial incident, etc. However, the increased dimensionality results in sparse data and overfitted models. This study proposes a new approach for efficient financial distress classification assessment by combining dimensionality reduction and machine learning techniques. The proposed framework aims to identify a subset of features leading to the minimization of the loss function describing the financial distress in an enterprise. During the study, 15 dimensionality reduction techniques with different numbers of features and 17 machine-learning models were compared. Overall, 1,432 experiments were performed using Lithuanian enterprise data covering the period from 2015 to 2022. Results revealed that the artificial neural network (ANN) model with 30 ranked features identified using the Random Forest mean decreasing Gini (RF_MDG) feature selection technique provided the highest AUC score. Moreover, this study has introduced a novel approach for feature extraction, which could improve financial distress classification models.

Enhanced Model Predictions through Principal Components and Average Least Squares-Centered Penalized Regression

We address the estimation of regression parameters for the ill-conditioned predictive linear model in this study. Traditional least squares methods often encounter challenges in yielding reliable results when there is multicollinearity. Therefore, we employ a better shrinkage method, average least squares-centered penalized regression (ALPR), as it offers a more efficient approach for handling multicollinearity than ridge regression. Additionally, we integrate ALPR with the principal component (PC) dimension reduction method for enhanced performance. We compared the proposed PCALPR estimation technique with existing ones for ill-conditioned problems through comprehensive simulations and real-life data analyses using the mean squared error. This integration results in superior model performance compared to other methods, highlighting the potential of combining dimensionality reduction techniques with penalized regression for enhanced model predictions.

Ensemble dynamics and information flow deduction from whole-brain imaging data.

The recent advancements in large-scale activity imaging of neuronal ensembles offer valuable opportunities to comprehend the process involved in generating brain activity patterns and understanding how information is transmitted between neurons or neuronal ensembles. However, existing methodologies for extracting the underlying properties that generate overall dynamics are still limited. In this study, we applied previously unexplored methodologies to analyze time-lapse 3D imaging (4D imaging) data of head neurons of the nematode Caenorhabditis elegans. By combining time-delay embedding with the independent component analysis, we successfully decomposed whole-brain activities into a small number of component dynamics. Through the integration of results from multiple samples, we extracted common dynamics from neuronal activities that exhibit apparent divergence across different animals. Notably, while several components show common cooperativity across samples, some component pairs exhibited distinct relationships between individual samples. We further developed time series prediction models of synaptic communications. By combining dimension reduction using the general framework, gradient kernel dimension reduction, and probabilistic modeling, the overall relationships of neural activities were incorporated. By this approach, the stochastic but coordinated dynamics were reproduced in the simulated whole-brain neural network. We found that noise in the nervous system is crucial for generating realistic whole-brain dynamics. Furthermore, by evaluating synaptic interaction properties in the models, strong interactions within the core neural circuit, variable sensory transmission and importance of gap junctions were inferred. Virtual optogenetics can be also performed using the model. These analyses provide a solid foundation for understanding information flow in real neural networks.

A novel hybrid machine learning model for rapid assessment of wave and storm surge responses over an extended coastal region

Storm surge and waves are responsible for a substantial portion of tropical and extratropical cyclones-related damages. While high-fidelity numerical models have significantly advanced the simulation accuracy of storm surge and waves, they are not practical to be employed for probabilistic analysis, risk assessment or rapid prediction due to their high computational demands. In this study, a novel hybrid model combining dimensionality reduction and data-driven techniques is developed for rapid assessment of waves and storm surge responses over an extended coastal region. Specifically, the hybrid model simultaneously identifies a low-dimensional representation of the high-dimensional spatial system based on a deep autoencoder (DAE) while mapping the storm parameters to the obtained low-dimensional latent space using a deep neural network (DNN). To train the hybrid model, a combined weighted loss function is designed to encourage a balance between DAE and DNN training and achieve the best accuracy. The performance of the hybrid model is evaluated through a case study using the synthetic data from the North Atlantic Comprehensive Coastal Study (NACCS) covering critical regions within New York and New Jersey. In addition, the proposed approach is compared with two decoupled models where the regression model is based on DNN and the reduction techniques are either principal component analysis (PCA) or DAE which are trained separately from the DNN model. High accuracy and computational efficiency are observed for the hybrid model which could be readily implemented as part of early warning systems or probabilistic risk assessment of waves and storm surge.

Multi-fidelity reduced-order surrogate modelling

High-fidelity numerical simulations of partial differential equations (PDEs) given a restricted computational budget can significantly limit the number of parameter configurations considered and/or time window evaluated. Multi-fidelity surrogate modelling aims to leverage less accurate, lower-fidelity models that are computationally inexpensive in order to enhance predictive accuracy when high-fidelity data are scarce. However, low-fidelity models, while often displaying the qualitative solution behaviour, fail to accurately capture fine spatio-temporal and dynamic features of high-fidelity models. To address this shortcoming, we present a data-driven strategy that combines dimensionality reduction with multi-fidelity neural network surrogates. The key idea is to generate a spatial basis by applying proper orthogonal decomposition (POD) to high-fidelity solution snapshots, and approximate the dynamics of the reduced states—time-parameter-dependent expansion coefficients of the POD basis—using a multi-fidelity long short-term memory network. By mapping low-fidelity reduced states to their high-fidelity counterpart, the proposed reduced-order surrogate model enables the efficient recovery of full solution fields over time and parameter variations in a non-intrusive manner. The generality of this method is demonstrated by a collection of PDE problems where the low-fidelity model can be defined by coarser meshes and/or time stepping, as well as by misspecified physical features.

2D MOF with multifunctionalized catalytic sites for one‐pot tandem synthesis of cyclic carbonates from olefins and CO2

AbstractOne‐pot tandem synthesis of cyclic carbonates from olefins and CO2 is becoming one of the most potential strategies to fix CO2 and yield high value chemicals. However, this reaction route is in need of high performing catalysts with multifunctionalized catalytic sites in single material which is still a big challenge in one‐pot community. In this term, we combined dimension reduction and surface modification strategies to fabricate a MOF/metal complex composite, that is, Zr‐BTB/PA‐Co. In this composite, the increased Lewis base sites is capable of boosting the adsorption and activation of CO2, which further acted as nucleophilic groups to activate epoxides, while the 2D matrix offered accessible intrinsic Lewis acid sites. These advantages synergistically led to superior catalytic performance of Zr‐BTB/PA‐Co for catalyzing one‐pot reaction of olefins and CO2 under mild conditions (80°C, 1 bar, 12 h and co‐catalyst/solvent free), and even impressive catalytic activity when using CO2 directly from air.

Predictive Deep Learning for High‐Dimensional Inverse Modeling of Hydraulic Tomography in Gaussian and Non‐Gaussian Fields

AbstractInverse modeling of hydraulic tomography (HT) is computationally expensive for estimating high‐dimensional hydrogeologic parameter fields. In this work, we develop a novel method called HT‐INV‐NN, which combines dimensionality reduction techniques with a predictive deep learning (DL) model to estimate high‐dimensional Gaussian and non‐Gaussian channel fields. The HT‐INV‐NN model consists of a predictor that directly learns the inverse process from hydraulic head measurements to latent variables of random fields, and a decoder that generates high‐dimensional parameter fields from predicted latent variables. For Gaussian spatially correlated fields, the decoder utilizes principal components derived from spatial covariance, and for non‐Gaussian channel fields, a generative adversarial network (GAN) is trained using generated realizations based on a training image (TI). The predictor is a deep neural network calibrated using the reference data obtained from HT forward simulations, which can be implemented in parallel. HT‐INV‐NN is successfully tested in multiple numerical experiments including steady‐state and transient HT for estimating Gaussian fields in 2D and 3D, as well as binary discontinuous or continuous non‐Gaussian channel fields. The training process is efficient, and the model structure demonstrates robustness for input data with perturbations. The model performance on multiple validation data sets are satisfactory when compared with other numerical and deep learning methods.

Combining dimensionality reduction methods with neural networks for realized volatility forecasting

Combining dimensionality reduction methods with neural networks for realized volatility forecasting

Self-updating digital twin of a hydrogen-powered furnace using data assimilation

Data assimilation, i.e., upgrading a numerical model by using experimental observations, is applied to adapt the performances of a simulation-based digital twin (DT) of a semi-industrial combustion furnace, based on available experimental data. More specifically, we rely on Kalman filter (KF) to adjust the prediction of our model by accounting for the underlying uncertainties. The DT is obtained by combining dimensionality reduction (through Proper Orthogonal Decomposition, POD) and regression (using Kriging) applied to Reynolds-averaged Navier–Stokes simulations of the furnace covering a three-dimensional design space, including both geometric and operational parameters. The experimental campaign concerns the measurement of the axial and radial profile of temperature inside the chamber and the NO concentrations at the outlet of the furnace, for a fuel mixture ranging from pure methane to pure hydrogen. Two types of KF algorithms are analyzed, i.e. the steady-state and the recursive ones. Both methodologies demonstrate improved DT performances, highlighting the significance of the Kalman gain in weighing the model’s prediction and measurement uncertainties. We also conduct a sensitivity analysis of data errors to reinforce this concept. The results of our study demonstrate the potential of data assimilation to build accurate and adaptive reduced-order models of realistic combustion systems.

PfaSTer: a machine learning-powered serotype caller for Streptococcus pneumoniae genomes.

Streptococcus pneumoniae (pneumococcus) is a leading cause of morbidity and mortality worldwide. Although multi-valent pneumococcal vaccines have curbed the incidence of disease, their introduction has resulted in shifted serotype distributions that must be monitored. Whole genome sequence (WGS) data provide a powerful surveillance tool for tracking isolate serotypes, which can be determined from nucleotide sequence of the capsular polysaccharide biosynthetic operon (cps). Although software exists to predict serotypes from WGS data, most are constrained by requiring high-coverage next-generation sequencing reads. This can present a challenge in respect of accessibility and data sharing. Here we present PfaSTer, a machine learning-based method to identify 65 prevalent serotypes from assembled S. pneumoniae genome sequences. PfaSTer combines dimensionality reduction from k-mer analysis with a Random Forest classifier for rapid serotype prediction. By leveraging the model's built-in statistical framework, PfaSTer determines confidence in its predictions without the need for coverage-based assessments. We then demonstrate the robustness of this method, returning >97 % concordance when compared to biochemical results and other in silico serotyping tools. PfaSTer is open source and available at: https://github.com/pfizer-opensource/pfaster.

RabbitTClust: enabling fast clustering analysis of millions of bacteria genomes with MinHash sketches

We present RabbitTClust, a fast and memory-efficient genome clustering tool based on sketch-based distance estimation. Our approach enables efficient processing of large-scale datasets by combining dimensionality reduction techniques with streaming and parallelization on modern multi-core platforms. 113,674 complete bacterial genome sequences from RefSeq, 455 GB in FASTA format, can be clustered within less than 6 min and 1,009,738 GenBank assembled bacterial genomes, 4.0 TB in FASTA format, within only 34 min on a 128-core workstation. Our results further identify 1269 redundant genomes, with identical nucleotide content, in the RefSeq bacterial genomes database.

Uncertainty Quantification for Multiphase Computational Fluid Dynamics Closure Relations with a Physics-Informed Bayesian Approach

Multiphase Computational Fluid Dynamics (MCFD) based on the two-fluid model is considered a promising tool to model complex two-phase flow systems. MCFD simulation can predict local flow features without resolving interfacial information. As a result, the MCFD solver relies on closure relations to describe the interaction between the two phases. Those empirical or semi-mechanistic closure relations constitute a major source of uncertainty for MCFD predictions. In this paper, we leverage a physics-informed uncertainty quantification (UQ) approach to inversely quantify the closure relations’ model form uncertainty in a physically consistent manner. This proposed approach considers the model form uncertainty terms as stochastic fields that are additive to the closure relation outputs. Combining dimensionality reduction and Gaussian processes, the posterior distribution of the stochastic fields can be effectively quantified within the Bayesian framework with the support of experimental measurements. As this UQ approach is fully integrated into the MCFD solving process, the physical constraints of the system can be naturally preserved in the UQ results. In a case study of adiabatic bubbly flow, we demonstrate that this UQ approach can quantify the model form uncertainty of the MCFD interfacial force closure relations, thus effectively improving the simulation results with relatively sparse data support.

Forecasting gold volatility with geopolitical risk indices

This paper tries to forecast gold volatility with multiple country-specific (GPR) indices and compares the role of combined prediction models and dimension reduction methods regarding the improvement of gold volatility prediction accuracy. For this purpose, GARCH-MIDAS model’s several extensions are used. We find firstly that most country-specific GPR indices have driving effects on gold volatility, and it makes sense to take forecast information from multiple country-specific GPR indices into account when forecasting gold volatility. The out-of-sample empirical results also indicate that the dimension reduction methods yield better predictions compared to the combined prediction models. In addition, dimension reduction technologies have excellent forecasting performance mainly during low gold volatility periods. Finally, our empirical findings are robust after changing the evaluation method, model settings, in-sample length and gold market.

Multistage classification identifies altered cortical phase- and amplitude-coupling in Multiple Sclerosis

Distinguishing groups of subjects or experimental conditions in a high-dimensional feature space is a common goal in modern neuroimaging studies. Successful classification depends on the selection of relevant features as not every neuronal signal component or parameter is informative about the research question at hand. Here, we developed a novel unsupervised multistage analysis approach that combines dimensionality reduction, bootstrap aggregating and multivariate classification to select relevant neuronal features. We tested the approach by identifying changes of brain-wide electrophysiological coupling in Multiple Sclerosis. Multiple Sclerosis is a demyelinating disease of the central nervous system that can result in cognitive decline and physical disability. However, related changes in large-scale brain interactions remain poorly understood and corresponding non-invasive biomarkers are sparse. We thus compared brain-wide phase- and amplitude-coupling of frequency specific neuronal activity in relapsing-remitting Multiple Sclerosis patients (n = 17) and healthy controls (n = 17) using magnetoencephalography. Changes in this dataset included both, increased and decreased phase- and amplitude-coupling in wide-spread, bilateral neuronal networks across a broad range of frequencies. These changes allowed to successfully classify patients and controls with an accuracy of 84%. Furthermore, classification confidence predicted behavioral scores of disease severity. In sum, our results unravel systematic changes of large-scale phase- and amplitude coupling in Multiple Sclerosis. Furthermore, our results establish a new analysis approach to efficiently contrast high-dimensional neuroimaging data between experimental groups or conditions.

Data-driven control of spatiotemporal chaos with reduced-order neural ODE-based models and reinforcement learning

Deep reinforcement learning (RL) is a data-driven method capable of discovering complex control strategies for high-dimensional systems, making it promising for flow control applications. In particular, the present work is motivated by the goal of reducing energy dissipation in turbulent flows, and the example considered is the spatiotemporally chaotic dynamics of the Kuramoto–Sivashinsky equation (KSE). A major challenge associated with RL is that substantial training data must be generated by repeatedly interacting with the target system, making it costly when the system is computationally or experimentally expensive. We mitigate this challenge in a data-driven manner by combining dimensionality reduction via an autoencoder with a neural ODE framework to obtain a low-dimensional dynamical model from just a limited data set. We substitute this data-driven reduced-order model (ROM) in place of the true system during RL training to efficiently estimate the optimal policy, which can then be deployed on the true system. For the KSE actuated with localized forcing (‘jets’) at four locations, we demonstrate that we are able to learn a ROM that accurately captures the actuated dynamics as well as the underlying natural dynamics just from snapshots of the KSE experiencing random actuations. Using this ROM and a control objective of minimizing dissipation and power cost, we extract a control policy from it using deep RL. We show that the ROM-based control strategy translates well to the true KSE and highlight that the RL agent discovers and stabilizes an underlying forced equilibrium solution of the KSE system. We show that this forced equilibrium captured in the ROM and discovered through RL is related to an existing known equilibrium solution of the natural KSE.

Active Learning Strategy for Surrogate-Based Quantile Estimation of Field Function

Uncertainty quantification is widely used in engineering domains to provide confidence measures on complex systems. It often requires to accurately estimate extreme statistics on computationally intensive black-box models. In case of spatially or temporally distributed model outputs, one valuable metric results in the estimation of extreme quantile of the output stochastic field. In this paper, a novel active learning surrogate-based method is proposed to determine the quantile of an unidimensional output stochastic process with a confidence measure. This allows to control the error on the estimation of a extreme quantile measure of a stochastic process. The proposed approach combines dimension reduction techniques, Gaussian process and an adaptive refinement strategy to enrich the surrogate model and control the accuracy of the quantile estimation. The proposed methodology is applied on an analytical test case and a realistic aerospace problem for which the estimation of a flight envelop is of prime importance for launch safety reasons in the space industry.

Subphenotyping of Mexican Patients With COVID-19 at Preadmission To Anticipate Severity Stratification: Age-Sex Unbiased Meta-Clustering Technique.

BackgroundThe COVID-19 pandemic has led to an unprecedented global health care challenge for both medical institutions and researchers. Recognizing different COVID-19 subphenotypes—the division of populations of patients into more meaningful subgroups driven by clinical features—and their severity characterization may assist clinicians during the clinical course, the vaccination process, research efforts, the surveillance system, and the allocation of limited resources.ObjectiveWe aimed to discover age-sex unbiased COVID-19 patient subphenotypes based on easily available phenotypical data before admission, such as pre-existing comorbidities, lifestyle habits, and demographic features, to study the potential early severity stratification capabilities of the discovered subgroups through characterizing their severity patterns, including prognostic, intensive care unit (ICU), and morbimortality outcomes.MethodsWe used the Mexican Government COVID-19 open data, including 778,692 SARS-CoV-2 population-based patient-level data as of September 2020. We applied a meta-clustering technique that consists of a 2-stage clustering approach combining dimensionality reduction (ie, principal components analysis and multiple correspondence analysis) and hierarchical clustering using the Ward minimum variance method with Euclidean squared distance.ResultsIn the independent age-sex clustering analyses, 56 clusters supported 11 clinically distinguishable meta-clusters (MCs). MCs 1-3 showed high recovery rates (90.27%-95.22%), including healthy patients of all ages, children with comorbidities and priority in receiving medical resources (ie, higher rates of hospitalization, intubation, and ICU admission) compared with other adult subgroups that have similar conditions, and young obese smokers. MCs 4-5 showed moderate recovery rates (81.30%-82.81%), including patients with hypertension or diabetes of all ages and obese patients with pneumonia, hypertension, and diabetes. MCs 6-11 showed low recovery rates (53.96%-66.94%), including immunosuppressed patients with high comorbidity rates, patients with chronic kidney disease with a poor survival length and probability of recovery, older smokers with chronic obstructive pulmonary disease, older adults with severe diabetes and hypertension, and the oldest obese smokers with chronic obstructive pulmonary disease and mild cardiovascular disease. Group outcomes conformed to the recent literature on dedicated age-sex groups. Mexican states and several types of clinical institutions showed relevant heterogeneity regarding severity, potentially linked to socioeconomic or health inequalities.ConclusionsThe proposed 2-stage cluster analysis methodology produced a discriminative characterization of the sample and explainability over age and sex. These results can potentially help in understanding the clinical patient and their stratification for automated early triage before further tests and laboratory results are available and even in locations where additional tests are not available or to help decide resource allocation among vulnerable subgroups such as to prioritize vaccination or treatments.

Identifying key features in reactive flows: A tutorial on combining dimensionality reduction, unsupervised clustering, and feature correlation

This study examines the capabilities of a data-driven workflow for automated key feature identification in reactive flows. The proposed approach aims at expediting the analysis of chemical engineering datasets by generating an automatic and explainable classification of regions showcasing distinct physics. The three main steps of this process, i.e., dimensionality reduction, unsupervised clustering, and feature correlation are discussed. A previously published framework based on these steps is used to compare against our proposed workflow, which employs different and more modern algorithms. The theoretical and practical differences between the previous and current algorithms are demonstrated in full. Overall, the key feature identification capability of the updated workflow is shown to be faster, more accurate, more robust, and closer to human intuition than previous methods. Throughout this study no substantial knowledge of machine learning is required from the reader. This makes this work also double up as a tutorial for researchers aiming at applying these algorithms.

Predictability of bull and bear markets: A new look at forecasting stock market regimes (and returns) in the US

The empirical literature of stock market predictability mainly suffers from model uncertainty and parameter instability. To meet this challenge, we propose a novel approach that combines dimensionality reduction, regime-switching models, and forecast combination to predict excess returns on the S&P 500. First, we aggregate the weekly information of 146 popular macroeconomic and financial variables using different principal component analysis techniques. Second, we estimate Markov-switching models with time-varying transition probabilities using the principal components as predictors. Third, we pool the models in forecast clusters to hedge against model risk and to evaluate the usefulness of different specifications. Our weekly forecasts respond to regime changes in a timely manner to participate in recoveries or to prevent losses. This is also reflected in an improvement of risk-adjusted performance measures as compared to several benchmarks. However, when considering stock market returns, our forecasts do not outperform common benchmarks. Nevertheless, they do add statistical and, in particular, economic value during recessions or in declining markets.