Numerous Datasets Research Articles

A multi-attention residual integrated network with enhanced fireworks algorithm for remote sensing image classification

This research examines a multi-attention residual integrated network with an enhanced fireworks algorithm for remote sensing image classification. Remote sensing (RS) picture classification is important for land cover mapping, environmental monitoring, and urban planning. Remote sensing image classification is important in earth observation since the military and commercial sectors have focused on it. Due to RS data's high complexity and limited labelled examples, classifying RS pictures is difficult. Deep Learning (DL) techniques have made great strides in RS image categorization, expanding this field's potential. This research introduces Multi-Attention Residual Integrated Network with Enhanced Fireworks Algorithm (MAR-EFA) to improve hyper spectral image identification. MARIN-EFA improves feature fusion and removes unneeded features to overcome technique constraints. The suggested method weights features using different attention models. These characteristics are then carefully extracted and integrated using a residual network. Final contextual semantic integration on deeply fused features is done with a Bi-LSTM network. Our population-based Enhanced Fireworks Algorithm (EFA) is inspired by fireworks' explosive performance and optimises MARIN parameters. Attention techniques and an improved optimisation algorithm improve performance over current systems. Numerous Eurosat dataset studies were assessed using various performance indicators. The simulation results show that MARIN-EFA outperforms current methods. The suggested technique shows promise for improving RS picture classification and allowing more accurate and reliable data categorization.

A Deep Learning-Based Diabetes Diagnosis Model on PIMA Image Dataset

Deep learning is a highly useful technique for the early identification of diabetes mellitus, according to the study done by numerous authors over the past few decades. By using pre-processing techniques on the dataset to get rid of various anomalies like over-fitting, under-fitting, redundancy, missing values, and non-significant features to make it more efficient for analysis, it is possible to increase the effectiveness of deep learning algorithms for diagnosing the disease. This work addresses the global problem of diabetes by exploring a revolutionary deep-learning method for early identification. Conventional convolutional neural network (CNN) models have drawbacks when used with numerical medical datasets, like this study's PIMA Indians Diabetes Database. The article suggests a technique for transforming numerical data into visual representations depending on feature relevance to get over this obstacle. This conversion makes it possible to use strong CNN models for diabetes early diagnosis. Classifying the created diabetic images after feeding them into CNN architectures that have already been trained on VGG16 and ResNet50. The promising outcomes with an accuracy of 97.19% demonstrate the possibility of the suggested strategy for improving diabetes detection and validating the effectiveness of diabetes imaging in obtaining an early diagnosis.

Heat transfer analysis for magnetohydrodynamic peristalsis of Reiner–Philippoff fluid: Application of an artificial neural network

Present communication explores a novel application of the computational intelligence technique, namely, the Levenberg–Marquardt scheme under a Backpropagated Neural Network (LM-BNN) to solve the mathematical model for the magnetohydrodynamic peristaltic transport of Reiner–Philippoff (R–Ph) pseudoplastic fluid considering the influences of Ohmic heating, mixed convection, and viscous dissipation through a symmetric channel. The R–Ph fluid model is used in this investigation to elucidate the non-Newtonian behavior of the fluid under consideration. The Reiner–Philippoff fluid model delineates the intricate relationship between stress and deformation rate within the fluid. There are a few studies available on the peristaltic transport of the Reiner–Philippoff fluid that do not incorporate Joule heating, mixed convection, and magnetic field effects. Therefore, a novel mathematical model is developed to employ an artificial neural network technique with a different approach that has not been examined before. The governing equations of the problem are simplified using long wavelength and low Reynolds number approximations, and the resulting system is numerically solved using the BVP4c scheme in MATLAB based on the shooting algorithm. Furthermore, a dataset is constructed through the BVP4c technique for the proposed LM-BNN, considering eight scenarios of peristaltic motion of the Reiner–Philippoff fluid model by varying the Bingham number, the Brinkman number, the Grashof number, the R–Ph fluid parameter, and the Hartmann number. The numerical dataset is divided into 15% for testing, 15% for training, and 70% for validation, which are utilized in LM-BNN to analyze the numerical solutions and Levenberg–Marquardt neural networks (LM-NNs) predicted results. The consistency and effectiveness of LM-BNN are validated through regression analysis, stresses at the wall, error histogram, correlation index, heat transfer, and mean squared error based fitness curves, which vary from 10−3→10−11. Variations in several flow parameters affecting temperature and velocity profiles are explained physically through graphs. Additionally, an analysis of heat transfer and stresses at the wall, including absolute errors, is provided through tables. The outcomes reveal that the improving Grashof number and the Hartmann number tend to increase the temperature profile. Tabular results indicate that rates of heat transfer improve when assigning higher values to the Hartmann number, the Brinkman number, and the Grashof number, whereas stresses at the wall decrease for the Reiner–Philippoff fluid parameter and the Bingham number. The error analysis of numerical simulations is a valuable step in determining whether the data obtained are reliable and accurate. In terms of absolute error, the disagreement between numerical solutions and those predicted by LM-NNs is approximately 10−5→10−11. It is clear from this error analysis that LM-NNs predicted results are consistent and reliable.

Crop Recommendation and Monitoring using AI

Abstract: Agriculture is crucial for India's economy, with over 50% relying on it for survival. Climate and weather variations pose risks to agriculture's health. AI can monitor crops by using machine learning methods. Crop monitoring includes Crop Recommendation, Weed Detection, Plant Disease Detection, Yield Prediction. The models are trained with Image and numerical datasets. A website will be developed to monitor crops and provide solutions. The optimal crop can be suggested based on the surrounding conditions by analysing important variables like composition of Nitrogen, Phosphorous and Potassium in the soil, its pH value, humidity, and rainfall using various models namely Gaussian Naive Bayes, Logistic Regression, Gradient boosting, Ensemble which fall under the domain of Machine Learning. ANN can be used for crop yield prediction. Weed and Plant disease can be detected using ResNet which can be utilized for deep neural networks. The intent of this project is to help farmers choose suitable crops, differentiate crops from weeds, detect diseases, provide remedies to protect crop. It enables to improve yield and productivity, Enhanced sustainability, Increased Profitability

Determination of the characteristic values of the undrained shear strength of organic soils according to Eurocode 7

The paper presents the methods of determining the characteristic value on the basis of the standards: PN-B-03020:1981, PN-EN 1997-1:2008, prEN 1997-1:2022-09 and Schneider formula. Determination of the characteristic value of the undrained shear strength τfu was carried out using statistical method on the basis of the prEN 1997-1:2022-09 standard and Schneider formula. The statistical calculations were based on the results of field vane tests carried out in organic subsoil of test embankment in Antoniny test site before loading and after the 2nd embankment stage. In order to determine the undrained shear strength τfu of organic soils from field vane tests, the measured values of shear strength τf v were corrected using the average values of correction factors μ = μ(lab) determined on the basis of triaxial compression, simple shear and triaxial extension tests. The analysis of the calculation results shows that with relatively numerous data sets, large values of the coefficient of variation Vx result in significantly lower characteristic values of τfu obtained according to prEN 1997-1:2022-09, compared to the values obtained according to the Schneider formula. In the case of few data sets, for which high values of the coefficient kn are obtained, with high values of the coefficient of variation Vx , the comparison of the values according to prEN 1997-1:2022-09 with the values obtained according to the Schneider formula shows the greatest differences.

Convolutional Autoencoder-Based medical image compression using a novel annotated medical X-ray imaging dataset

With the ongoing development of machine learning techniques, it is now necessary to train and evaluate these algorithms to have access to high-quality medical X-ray datasets. This study unfolds on two critical axes within the realm of medical imaging. We introduce the proposed Medical X-ray Imaging Dataset (MXID), a meticulously curated resource featuring images spanning 18 body parts. It has been refined to include a classification based on body type within each gender category. This dataset addresses the limitations of existing datasets by offering comprehensive coverage, precise annotations, and optimal image quality. While numerous datasets often fall short in terms of encompassing a broad range of anatomical regions, particularly those that are designed for multi-body component analysis. Our primary contribution lies in bridging the gap in available resources, providing a foundation of developing robust machine learning algorithms in medical imaging. Moreover, we extend our investigation to the integration of the MXID dataset into the domain of medical image compression, evaluating the performance of techniques such as Principal Component Analysis (PCA), K-means clustering, Convolutional Neural Network (CNN), Deep Convolutional Autoencoders (DCAEs), Autoencoders (AEs), and Variational Autoencoder (VAE), using two scenarios, one is for single image compression, and the other is for MXID dataset, which offer a unique perspective that enables a comprehensive assessment of these techniques across diverse anatomical regions. Our study contributes to advancing machine learning and deep learning applications in medical imaging, emphasizing the creation of a dataset and its integration into image compression.

UniCell: Universal Cell Nucleus Classification via Prompt Learning

The recognition of multi-class cell nuclei can significantly facilitate the process of histopathological diagnosis. Numerous pathological datasets are currently available, but their annotations are inconsistent. Most existing methods require individual training on each dataset to deduce the relevant labels and lack the use of common knowledge across datasets, consequently restricting the quality of recognition. In this paper, we propose a universal cell nucleus classification framework (UniCell), which employs a novel prompt learning mechanism to uniformly predict the corresponding categories of pathological images from different dataset domains. In particular, our framework adopts an end-to-end architecture for nuclei detection and classification, and utilizes flexible prediction heads for adapting various datasets. Moreover, we develop a Dynamic Prompt Module (DPM) that exploits the properties of multiple datasets to enhance features. The DPM first integrates the embeddings of datasets and semantic categories, and then employs the integrated prompts to refine image representations, efficiently harvesting the shared knowledge among the related cell types and data sources. Experimental results demonstrate that the proposed method effectively achieves the state-of-the-art results on four nucleus detection and classification benchmarks. Code and models are available at https://github.com/lhaof/UniCell

Amalgamating Multi-Task Models with Heterogeneous Architectures

Multi-task learning (MTL) is essential for real-world applications that handle multiple tasks simultaneously, such as selfdriving cars. MTL methods improve the performance of all tasks by utilizing information across tasks to learn a robust shared representation. However, acquiring sufficient labeled data tends to be extremely expensive, especially when having to support many tasks. Recently, Knowledge Amalgamation (KA) has emerged as an effective strategy for addressing the lack of labels by instead learning directly from pretrained models (teachers). KA learns one unified multi-task student that masters all tasks across all teachers. Existing KA for MTL works are limited to teachers with identical architectures, and thus propose layer-to-layer based approaches. Unfortunately, in practice, teachers may have heterogeneous architectures; their layers may not be aligned and their dimensionalities or scales may be incompatible. Amalgamating multi-task teachers with heterogeneous architectures remains an open problem. For this, we design Versatile Common Feature Consolidator (VENUS), the first solution to this problem. VENUS fuses knowledge from the shared representations of each teacher into one unified generalized representation for all tasks. Specifically, we design the Feature Consolidator network that leverages an array of teacher-specific trainable adaptors. These adaptors enable the student to learn from multiple teachers, even if they have incompatible learned representations. We demonstrate that VENUS outperforms five alternative methods on numerous benchmark datasets across a broad spectrum of experiments.

Abstract 6475: Leveraging real-world pancreatic cancer datasets to drive drug discovery and patient health

Abstract Pancreatic cancer is the third leading cause of cancer-related death in the United States. Current therapeutic options offer a dismal overall survival with the 5-year survival at just ~12%. Analysis of the clinical and molecular underpinnings of pancreatic cancer is critical to developing both early detection methodologies as well as novel therapeutic options. The aggressiveness and deadly nature of this disease warranted the development of a data repository and analysis system dedicated to pancreatic cancer data. The Pancreatic Cancer Action Network’s (PanCAN) SPARK platform is a cloud-based data and analytics platform, powered by Velsera, that integrates real-world patient health data from PanCAN research initiatives and accelerates research by making pancreatic cancer data easier to access and use. Encompassing clinical, imaging, and genomics data from over 600 patients with pancreatic cancer within PanCAN’s Know Your Tumor ® (KYT) precision medicine service, the SPARK platform connects with petabytes of publicly available cancer data via the Cancer Genomics Cloud (CGC), also powered by Velsera. The CGC is part of NCI’s Cancer Research Data Commons (CRDC), a cloud-based data science infrastructure that connects data with analytics tools to allow researchers to share, integrate, analyze, visualize, and drive scientific discovery. Here, we demonstrate the application of these datasets by providing a case study demonstrating how to combine and enrich data to accelerate pancreatic cancer research. Currently, the genomic and proteomic data available on CRDC amounts to 402 and 304 cases of pancreatic tumor samples, respectively. We will use the capabilities of the SPARK and CGC platforms, which provide ready-to-use tools for multi-omics analysis that require no coding knowledge. Using the KYT and CRDC open-access pancreatic cancer data, we aim to demonstrate how to perform integrated analysis of data from diverse scientific domains, and share with collaborators all in one space, streamlining and increasing the potential for new scientific discoveries. Further expansion of the PanCAN and CGC datasets will undoubtedly provide a more comprehensive understanding of pancreatic cancer tumor biology. SPARK and CGC’s cloud based computation infrastructure, along with numerous available cancer datasets and easy-to-use multi-omics data processing workflows and data analytic tools will be instrumental in this process. Citation Format: Kawther Abdilleh, Zelia Worman, Rowan Beck, Cera Fisher, Divya Sain, Jack DiGiovanna, Lynn Matrisian, Sudheer Doss, Brandi Davis-Dusenbery. Leveraging real-world pancreatic cancer datasets to drive drug discovery and patient health [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 6475.

Abstract 6867: TWEAK-Fn14 signaling in glioblastoma-associated myeloid cells promotes tumor invasion via CCL5-CCR5 axis

Abstract Glioblastoma remains difficult to treat with immunotherapy due to the immunosuppressive tumor microenvironment and the ability of glioma cells to invade diffusely into normal brain parenchyma. While it is known that glioblastoma-associated myeloid cells (GAMs) comprise a large proportion of the tumor microenvironment and can suppress adaptive and innate immune responses through various mechanisms, detailed mapping of myeloid lineages in glioblastoma is still missing. Further characterization of pro-tumoral myeloid-lineages in glioblastoma is important for identifying myeloid-based drug targets that may synergize with T cell-based therapies and produce a more meaningful anti-tumor immune response. Towards this end, we examined numerous scRNAseq datasets derived from glioblastoma patients and identified expression of TNFRSF12A, the gene encoding the TNF- family receptor, Fn14, in a subset of GAMs enriched for expression of cytokines such as CCL3, CCL5, IL-6, and IL-8. This result is intriguing since TWEAK (TNFSF12), the ligand for Fn14, is abundantly expressed within numerous patient-derived glioblastoma cell lines, namely GBM6, GBM8, GBM22, and GBM39. To validate Fn14 protein expression in GAMs, we isolated murine bone-marrow derived macrophages and treated them with glioma-conditioned media, which induced Fn14 gene expression. We then conducted in vitro experiments utilizing the human microglia line HMC3, the murine microglia line SIM-A9, and primary murine bone-marrow-derived macrophages to assess the effects of Fn14 activation in myeloid cells. We found that treatment of Fn14+ myeloid cells with TWEAK ligand upregulates CCL3, CCL5, IL-6, and IL-8 expression. We tested the conditioned media generated from TWEAK-treated Fn14+ myeloid cell lines and found that it promotes glioblastoma invasion in a CCL5-dependent manner. This aligns with previous reports demonstrating the CCL5-CCR5 signaling in glioma cells promotes invasion through calcium-dependent matrix metalloproteinase 2. To assess the spatial distribution of this myeloid subpopulation, we performed RNAscope hybridization against AIF1, TMEM119, CCR5, CCL5, and Fn14 transcripts on sections of murine gliomas generated using the RCAS/tv-a model of human glioblastoma. We found an abundance of Fn14+ plus AIF1+ as well as Fn14+plus TMEM119+ myeloid cells throughout the tumor core and rim of murine tumors and identified CCL5 expression in pseudo-palisades bordering necrotic regions. Our results to date suggest that Fn14 expression in GAMs can promote the invasion of tumor cells into normal brain parenchyma by release of CCL5. Currently, we are utilizing orthotopic patient-derived xenograft models co-implanted with Fn14+/- myeloid cells to test the hypothesis that TWEAK-Fn14 signaling in GAMs provides strong pro-tumoral signals in the glioblastoma microenvironment. Citation Format: Angad Beniwal, Matthew Dufault, Pranjali Kanvinde, David Nascari, Ryan Eghlimi, Adarsha Malla, Dolores Hambardzumyan, Henrique Borges da Silva, Graeme Woodworth, Jeffrey Winkles, Nhan Tran. TWEAK-Fn14 signaling in glioblastoma-associated myeloid cells promotes tumor invasion via CCL5-CCR5 axis [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 6867.

Abstract 6210: The cancer omics and drug experimental response dataset (CODERData): A harmonized benchmark dataset for machine learning models of drug response prediction

Abstract Background: Determining a patient’s response to a specific therapy is a vital step in developing personalized cancer treatment. Personalized treatment relies on two key technological advancements: algorithms tailored to each patient’s molecular profiles, and comprehensive datasets detailing the effects of patient diversity on drug response. Despite recent advancements in machine learning-based models, each new algorithm requires curation of numerous datasets, which poses a significant burden to researchers who need to evaluate and compare various algorithms. Currently, datasets include extensive genomics, transcriptomics, proteomics and metabolomics measurements along with drug response data (i.e., cell viability) for each cell line, collected through the Cancer Cell Line Encyclopedia (CCLE) and other initiatives. Additional omics data is available for tumor organoids and patient samples generated by consortia such as Human Cancer Model Initiative (HCMI) and the Clinical Proteomic Tumor Analysis Consortium (CPTAC). However, to date there has been little public effort to annotate and harmonize these data across repositories. Methods: To address this need, we developed the Python package coderdata that collates the most up-to-date cancer omics and drug response data. We curated four public cancer datasets, establishing reproducible pipelines that standardize data retrieval, formatting, and integration so the datasets can be built on a cluster if needed. The datasets undergo bi-monthly synchronization with FigShare, ensuring the data is up to date and available. Scientists can use coderdata to retrieve and format data on the command line interface as well as within Python scripts. Results: Coderdata is a straightforward Python package that will enable the development and benchmarking of diverse machine learning applications in cancer research. At its core, it represents reproducible data harmonization that enables heterogeneous datasets to be analyzed in bulk. This Python package downloads and standardizes data from multiple consortia-related resources including HCMI, CPTAC, CCLE, and BeatAML, together representing 4931 multi-omics samples from across 250+ cancer types. These datasets include cancer data metrics such as copy number, mutation, transcriptomics, proteomics, miRNA, methylation, and drug response data. Conclusions: The applications of machine learning in cancer biology are hampered by the distributed nature of existing datasets. As such, the collection and standardization of data by coderdata substantially reduces the time investment required for data curation in cancer research. Direct access to benchmark multi-omics and drug response datasets enables scientists to focus on algorithm development for tumor drug response prediction, potentially accelerating the discovery of therapeutic strategies. Citation Format: Jeremy Jacobson, Sydney Schwartz, M. Ryan Weil, Neeraj Kumar, Sara Gosline. The cancer omics and drug experimental response dataset (CODERData): A harmonized benchmark dataset for machine learning models of drug response prediction [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 6210.

Effect of structural parameters on compression performance of autoclaved aerated concrete: Simulation and machine learning

Autoclaved aerated concrete (AAC) has broad applications in civil engineering due to its lightweight, thermal and sound insulation. However, the relation between the complex random porous structures and the compression performances of AAC is complex, limiting the optimization and application of AAC. In this study, finite element simulation was implemented and validated based on the experimental results. The simulation results showed that the compressive strength of AAC increased by 31.7, 45.8, and 134% with the porosity decreasing from 70% to 40%, the average pore diameter decreasing from 1.5 to 0.5 mm, and the pore connectivity decreasing from 50% to 0, respectively. Then, based on the numerical dataset, integrated machine learning methods were implemented to rapidly predict the compressive properties of AAC and analyze the main factors affecting the compressive properties. The ML results showed that the CatBoost model had the best predictive performance based on the small dataset of simulation results, with an average relative error of 18% and 7% for compressive strength and modulus of AAC, respectively. The component modulus was the most important feature for predicting the compressive strength and modulus of AAC.

A Global Seawater Density Distribution Model Using a Convolutional Neural Network.

Seawater density is an important physical property in oceanography that affects the accuracy of calculations such as gravity fields and tidal potentials and the calibration of acoustic and optical oceanographic sensors. In related studies, constant density values are frequently used, which can introduce significant errors. Therefore, this study employs a basic convolutional neural network model to construct a comprehensive model showing the seawater density distribution across the globe. The model takes into account depth, latitude, longitude, and month as inputs. Numerous real seawater datasets were used to train the model, and it has been shown that the model has an absolute mean error and root mean square error of less than 1 kg/m3 in 99% of the test set samples. The model effectively demonstrates the influence of input parameters on the distribution of seawater density. In this paper, we present a newly developed global model for distributing seawater density which is both comprehensive and accurate, surpassing previous models. The utilization of the model presented in this paper for estimating seawater density can minimize errors in theoretical ocean models and serve as a foundation for designing and analyzing ocean exploration systems.

Investigating measurement invariance in PIRLS 2021 across English and isiZulu language groups

Background: PIRLS 2021 results revealed that South African Grade 4 learners performed significantly lower compared to other countries in reading comprehension and that they did not reach the standardised international mean score of 500. It was also evident from the results that English learners performed relatively higher than isiZulu learners.Objective: This study was initiated to investigate whether possible measurement invariance can help explain the difference in the scores of English and isiZulu learners. Measurement invariance is concerned with the metric and functional equivalence of the items in the two languages examined.Method: A quantitative approach was utilised to reanalyse the PIRLS 2021 numerical data sets. Descriptive statistics were used to analyse the mean and raw scores. Rasch analysis was conducted to look for any Differential Item Function (DIF) in the items.Results: Statistics revealed that there is a significant difference in the mean scores of the two languages. Analysis of raw scores provided evidence that some of the items lacked metric equivalence. Rasch analysis displayed that some of the items lacked functional equivalence.Conclusions: The significant difference found in the scores implies that Grade 4 learners need more attention mastering the skill of reading literacy. The lack of metric and functional equivalence suggests that items in the passages need to be investigated and improved.Contributions: This study has contributed to the International Large-Scale Assessment literature specifically relating to the equivalence of ILSA across languages. A significant gap in the learners’ scores was identified and measurement invariance helped explain the gap in the learners’ reading scores.

Deep Learning Ghost Polarimetry

The first application of neural networks in the problem of ghost polarimetry is reported. The proposed approach has enabled the reconstruction of the spatial distribution of object anisotropy in ghost polarimetry. The deep neural network processes a set of intensity correlation functions measured in various polarization states of classical light and reconstructs, point-by-point, the distribution of the type of anisotropy. In this work we use a numerical dataset. We investigated the applicability of the developed network for objects whose properties are determined by linear/circular amplitude/phase anisotropy. The probability of correctly predicting the type of anisotropy exceeds 95% according to the F1-score metric.

Modified stochastic diffusion particle tracking model driven by skew Brownian motion: Analysis of sediment particle motion in turbulent flow under the effects of ejection and sweep.

Turbulent bursting events have been classified into outward interactions (Q1), ejections (Q2), inward interactions (Q3), and sweeps (Q4) in various studies. Ejections (Q2) and sweeps (Q4) have been identified as significant contributors to time consumption, momentum flux, and sediment flux. Additionally, research has shown that the distribution of these events varies nonuniformly at different bed elevations. Despite extensive investigations into the nonuniform distribution of turbulent bursting events, their impact on sediment transport has been rarely explored. In this work, we developed a modified stochastic diffusion particle tracking model (SD-PTM) driven by skew Brownian motion (SBM) using the stochastic Lagrangian approach to scrutinize sediment particle movement in turbulent flows. The model incorporates turbulent characteristics derived from a direct numerical simulation dataset, allowing for a comprehensive analysis of sediment particle dynamics. Moreover, the proposed model accounts for the nonuniform spatial distribution of ejection and sweep events, as well as the particle movement direction during these events. Numerical simulations of the model were conducted to trace sediment particle trajectories in the streamwise and vertical directions. The analysis of sediment transport involved calculating the variance of particle trajectories to examine anomalous diffusion. The model's performance was evaluated by comparing it with flow velocity and sediment concentration profiles obtained from measurements in previous studies. In conclusion, our study suggests that the motion of sediment particles in turbulent flow can be thoroughly investigated under extreme flow conditions using the modified SD-PTM driven by SBM.

Data-driven modeling and forecasting of chaotic dynamics on inertial manifolds constructed as spectral submanifolds.

We present a data-driven and interpretable approach for reducing the dimensionality of chaotic systems using spectral submanifolds (SSMs). Emanating from fixed points or periodic orbits, these SSMs are low-dimensional inertial manifolds containing the chaotic attractor of the underlying high-dimensional system. The reduced dynamics on the SSMs turn out to predict chaotic dynamics accurately over a few Lyapunov times and also reproduce long-term statistical features, such as the largest Lyapunov exponents and probability distributions, of the chaotic attractor. We illustrate this methodology on numerical data sets including delay-embedded Lorenz and Rössler attractors, a nine-dimensional Lorenz model, a periodically forced Duffing oscillator chain, and the Kuramoto-Sivashinsky equation. We also demonstrate the predictive power of our approach by constructing an SSM-reduced model from unforced trajectories of a buckling beam and then predicting its periodically forced chaotic response without using data from the forced beam.

Trustworthy multi-view clustering via alternating generative adversarial representation learning and fusion

Multi-view clustering (MVC) has attached extensive attention as it provides an effective approach to deal with the unlabeled data in real-world applications. To enjoy the strong feature extraction capacity of deep learning, traditional shallow MVC methods are further extended to deep version. Though achieve superiorities in many fields, most existing deep MVC models are still limited by their lacking assurance of the trustworthiness and robustness. This drawback also leads to the possible poor clustering performance. To tackle this drawback, we propose an alternating generative adversarial learning strategy based network, termed as Alternating Generative Adversarial Representation Learning (AGARL) network for multi-view clustering, aiming to drive different views to fall into the same semantic space for the further alignment and information fusion. Various from the traditional generative adversarial learning strategy, the proposed alternating adversarial representation learning strategy successes in further aligning the distributions of the view-specific representations and exploring the cross-view consistency credited by its setting each view as the reference alternately. Then a consensus representative latent representation can be learned through an attention-like sub-network for all views. Finally, a self-supervised clustering module is employed in the proposed network to guide the learning of cluster assignment. In addition, comprehensive experiments on numerous benchmark MVC datasets have been performed to demonstrate the effectiveness of the proposed network compared to other state-of-the-art methods.

Semi-analytical Model to Predict Dynamic Capillary Pressure–Saturation Relationship for Flows in Heterogeneous Porous Media

The capillary pressure defines pressure difference between non-wetting and wetting fluids. The capillary pressure is part of the flow governing equations, and its definition can have a profound impact on the nature of fluids displacement in a multiphase flow environment. Conventionally, capillary pressure–saturation relationships are determined under equilibrium conditions which signify that all the fluid–fluid interfaces that exist at the pore scale maintain a static configuration at a certain instant in time. However, there exist experimental and numerical evidences that state that the dynamic nature of fluid flows indeed plays a prominent role in defining the trends of the capillary pressure–saturation relationships. In this paper, we develop a first of a kind semi-analytical model to predict the capillary pressure–saturation curves during drainage displacement by integrating the dynamics of fluid flow based on fundamental laws of fluid mechanics. The proposed semi-analytical model can potentially be incorporated into existing multiphase flow simulators to rapidly compute the capillary pressure at various saturations of the flow medium under dynamic flow conditions. The presented semi-analytical model has been validated against experimental and numerical data sets available in the literature at various flow conditions and considering different sets of fluid properties. We noticed a satisfactory match of the results predicted by the proposed semi-analytical model against the literature data. After performing a holistic sensitivity analysis, we notice that the properties of the porous medium, and the fluid–solid interactions play a significant role in defining the trends of the capillary pressure–saturation curves.

Acquisition of representative objects and attribute reductions based on generalized decisions of dominance-based rough set approach

Dominance principles play a pivotal role in decision makings, in which the superior the object performs on criteria, the higher the object will be assigned by decision attributes. In practical issues, the result of decision makers is influenced significantly because of the existence on redundant objects and attributes. However, little work studies the inconsistency of dominance principles caused by redundant objects. The proposed generalized decision aims to intuitively reveal the consistency of dominance principles. Since then, considerable attention has been devoted to accelerating attribute reductions in dominance-based rough sets. But there has been limited discussion on the semantic explanation of generalized decisions. This paper addresses this gap by introducing a graded information granule that not only explains the consistency of dominance principles but also facilitates the partitioning of the universe. Based on this foundation, several theoretical properties are thoroughly investigated. Moreover, we acquire a representative object set by applying a threshold to filter out objects with an extreme imbalance of dominance principles, and a final compact data set is generated with the best threshold by a particular parameter selection mechanism. Additionally, we design an attribute reduction method based on graded information granules using the compact data set. Furthermore, we employ numerical data sets to illustrate the effectiveness and efficiency of our method by running time and importance of reducts performed from the perspective of dominance-based rough sets and machine learning. Finally, the practicability of compact data sets is explained from the viewpoint of decision rules in dominance-based rough set approach.