Model Data Research Articles

Unit Maxwell-Boltzmann Distribution and Its Application to Concentrations Pollutant Data

In the vast statistical literature, there are numerous probability distribution models that can model data from real-world phenomena. New probability models, nevertheless, are still required in order to represent data with various spread behaviors. It is a known fact that there is a great need for new models with limited support. In this study, a flexible probability model called the unit Maxwell-Boltzmann distribution, which can model data values in the unit interval, is derived by selecting the Maxwell-Boltzmann distribution as a base-line model. The important characteristics of the derived distribution in terms of statistics and mathematics are investigated in detail in this study. Furthermore, the inference problem for the mentioned distribution is addressed from the perspectives of maximum likelihood, method of moments, least squares, and maximum product space, and different estimators are obtained for the unknown parameter of the distribution. The derived distribution outperforms competitive models according to different fit tests and information criteria in the applications performed on four actual air pollutant concentration data sets, indicating that it is an effective model for modeling air pollutant concentration data.

A Recent Development of a Network Approach to Assessment Data: Latent Space Item Response Modeling for Intelligence Studies.

This article aims to provide an overview of the potential advantages and utilities of the recently proposed Latent Space Item Response Model (LSIRM) in the context of intelligence studies. The LSIRM integrates the traditional Rasch IRT model for psychometric data with the latent space model for network data. The model has person-wise latent abilities and item difficulty parameters, capturing the main person and item effects, akin to the Rasch model. However, it additionally assumes that persons and items can be mapped onto the same metric space called a latent space and distances between persons and items represent further decreases in response accuracy uncaptured by the main model parameters. In this way, the model can account for conditional dependence or interactions between persons and items unexplained by the Rasch model. With two empirical datasets, we illustrate that (1) the latent space can provide information on respondents and items that cannot be captured by the Rasch model, (2) the LSIRM can quantify and visualize potential between-person variations in item difficulty, (3) latent dimensions/clusters of persons and items can be detected or extracted based on their latent positions on the map, and (4) personalized feedback can be generated from person-item distances. We conclude with discussions related to the latent space modeling integrated with other psychometric models and potential future directions.

Li-ion battery capacity prediction using improved temporal fusion transformer model

Lithium-ion (Li-ion) batteries have near-zero energy emissions and provide power to various devices, such as automobiles and portable equipment. The strategy predicts the capacity of Li-ion in advance and can also help arrange maintenance tasks. To improve state of health (SOH) and remaining useful life (RUL) prediction accuracy, we propose an improved temporal fusion transformer (ITFT) method based on bidirectional long short-term memory (Bi-LSTM) encoder-decoder layer replacing the long short-term memory for probabilistic online RUL prediction. A novel interpretable hyperparameter tuning method called Bayesian optimization based on tree-structure Parzen estimator (TPE) is coupled with a unique ITFT model to improve RUL prediction accuracy. Furthermore, we consider the effects of keen-onset to establish the starting point of our training. The root mean square error for four batteries using the proposed model for the test data are 0.0018, 0.0019, 0.0013, and 0.0025, respectively, which outperforms the other models, with an improvement accuracy rate above 25%. The proposed model SOH results indicate that our proposed approach outperforms some previously published methods. Our online RUL prediction demonstrates relative errors of 1.18%, 1.54%, 1.06%, 2.70%, 0.67%, 2%, 3.90%, 0%, and 3.08% for nine batteries, respectively. These results for SOH and RUL predictions emphasize the excellent performance of our proposed method.

Privacy Protection Based on Federated Learning

With the development of artificial intelligence technology, more and more fields will collect relevant user data, and provide users with a better experience through data analysis. But there are also risks involved in the process of data collection, namely how to protect personal privacy data. To address this issue, this study combined differential privacy, convolutional neural networks, and federated averaging algorithms to construct a privacy protection model. The study first utilized the federated average algorithm to handle data imbalance, ensuring that each analyzed data is in a balanced state. Then, based on of data balancing, a new algorithm model was constructed using differential privacy and convolutional neural networks. Finally, it utilized a number of public datasets to verify the role of the model in privacy protection. The results showed that the model can achieve recognition accuracy of 97.27% and 93.15%, respectively , for data under the influence of privacy budget and relaxation factor. Meanwhile, the classification accuracy of the model for data can reached 95.31%, with a regression error of 9.03%. When the local iteration number of the device was 30, the testing accuracy can reached 95.28%. This indicates that methods on the grounds of federated averaging algorithm and differential privacy can maintain the accuracy of the model while protecting user privacy. The application research of models has strong practical significance.

Dynamical Model of Rotation and Orbital Coupling for Deimos

This paper introduces a novel dynamical model, building upon the existing dynamical model for Deimos in the current numerical ephemerides, which only encompasses the simple libration effects of Deimos. The study comprehensively incorporates the rotational dynamics of Deimos influenced by the torque exerted by the major celestial bodies (Mars, the Sun) in the solar system within the inertial space. Consequently, a full dynamical model is formulated to account for the complete coupling between the rotation and orbit of Deimos. Simultaneously, employing precision orbit determination methods used for artificial satellites, we develop an adjustment model for fitting data to the complete model. The 12-order Adams–Bashforth–Moulton (ABM) integration algorithm is employed to synchronously integrate the 12 state variables of the full model to obtain the orbit of Deimos.The difference in the orbits obtained by integrating the full model over a period of 10 years and those obtained by the simplified model is at the order of 10 km. After precise orbit determination, this difference decreases to below 100 m, so numerical simulation results indicate that the full dynamical model and adjustment model are stable and reliable. Simultaneously, the integration of the Deimos third-order gravity field in the full model over a 10-year period induces only meter-level positional changes. This suggests that when constructing the complete model, the utilization of a second-order gravity field alone is sufficient. Compared to the simple model, the polar axis of Deimos in the inertial space exhibits a more complex oscillation in the full model. Additionally, the full model calculates that the minimum moment of inertia principal axis of Phobos has an amplitude of approximately 0.5 degrees in the longitude direction and does not exceed 2 degrees in the latitude direction. This work further advances the current dynamical model for Deimos and establishes the foundational model for the generation of a new set of precise numerical ephemerides for Deimos.

Deep partially linear cox model for current status data.

Deep learning has continuously attained huge success in diverse fields, while its application to survival data analysis remains limited and deserves further exploration. For the analysis of current status data, a deep partially linear Cox model is proposed to circumvent the curse of dimensionality. Modeling flexibility is attained by using deep neural networks (DNNs) to accommodate nonlinear covariate effects and monotone splines to approximate the baseline cumulative hazard function. We establish the convergence rate of the proposed maximum likelihood estimators. Moreover, we derive that the finite-dimensional estimator for treatment covariate effects is $\sqrt{n}$-consistent, asymptotically normal, and attains semiparametric efficiency. Finally, we demonstrate the performance of our procedures through extensive simulation studies and application to real-world data on news popularity.

Garlic Origin Traceability and Identification Based on Fusion of Multi-Source Heterogeneous Spectral Information.

The chemical composition and nutritional content of garlic are greatly impacted by its production location, leading to distinct flavor profiles and functional properties among garlic varieties from diverse origins. Consequently, these variations determine the preference and acceptance among diverse consumer groups. In this study, purple-skinned garlic samples were collected from five regions in China: Yunnan, Shandong, Henan, Anhui, and Jiangsu Provinces. Mid-infrared spectroscopy and ultraviolet spectroscopy were utilized to analyze the components of garlic cells. Three preprocessing methods, including Multiple Scattering Correction (MSC), Savitzky-Golay Smoothing (SG Smoothing), and Standard Normalized Variate (SNV), were applied to reduce the background noise of spectroscopy data. Following variable feature extraction by Genetic Algorithm (GA), a variety of machine learning algorithms, including XGboost, Support Vector Classification (SVC), Random Forest (RF), and Artificial Neural Network (ANN), were used according to the fusion of spectral data to obtain the best processing results. The results showed that the best-performing model for ultraviolet spectroscopy data was SNV-GA-ANN, with an accuracy of 99.73%. The best-performing model for mid-infrared spectroscopy data was SNV-GA-RF, with an accuracy of 97.34%. After the fusion of ultraviolet and mid-infrared spectroscopy data, the SNV-GA-SVC, SNV-GA-RF, SNV-GA-ANN, and SNV-GA-XGboost models achieved 100% accuracy in both training and test sets. Although there were some differences in the accuracy of the four models under different preprocessing methods, the fusion of ultraviolet and mid-infrared spectroscopy data yielded the best outcomes, with an accuracy of 100%. Overall, the combination of ultraviolet and mid-infrared spectroscopy data fusion and chemometrics established in this study provides a theoretical foundation for identifying the origin of garlic, as well as that of other agricultural products.

Analysis of the relationship between financial policy-competitiveness capacity - economic growth: A study of economic transformation in Vietnam

This study aims to examine the relationship between fiscal policy, provincial competitiveness and Vietnam's economic growth. The study uses the VECM model for time series data including the factors Total tax revenue of the country, Growth of the expanded money supply represents fiscal policy; Provincial competitiveness and Vietnam's economic growth in the period 1990-2021. The study has provided additional empirical evidence on the relationship between financial policy, provincial competitiveness and economic growth in the context of selecting Vietnam. The research results show that financial policies have affected economic growth in localities in the third period with an impact of 3%. In comparison, provincial competitiveness has affected economic growth right in the 1st period with a large impact of 18%. The change of economic growth has an impact in the same direction with provincial-level financial and energy competitiveness policies. From this result, the authors propose implications to limit the lag in the implementation of financial policies.

Enhanced degradation of Direct Red 80 dye via Fenton-like process mediated by cobalt ferrite: generated superoxide radicals and singlet oxygen.

Azo dyes, widely used in the textile industry, contribute to effluents with significant organic content. Therefore, the aim of this work was to synthesize cobalt ferrite (CoFe2O4) using the combustion method and assess its efficacy in degrading the azo dye Direct Red 80 (DR80). TEM showed a spherical structure with an average size of 33 ± 12nm. Selected area electron diffraction and XRD confirmed the presence of characteristic crystalline planes specific to CoFe2O4. The amount of Co and Fe metals were determined by ICP-OES, indicating an n(Fe)/n(Co) ratio of 2.02. FTIR exhibited distinct bands corresponding to Co-O (455cm-1) and Fe-O (523cm-1) bonds. Raman spectroscopy detected peaks associated with octahedral and tetrahedral sites. For the first time, the material was applied to degrade DR80 in an aqueous system, with the addition of persulfate. Consistently, within 60min, these trials achieved nearly 100% removal of DR80, even after the material had undergone five cycles of reuse. The pseudo-second-order model was found to be the most fitting model for the experimental data (k2 = 0.07007 L mg-1min-1). The results strongly suggest that degradation primarily occurred via superoxide radicals and singlet oxygen. Furthermore, the presence of UV light considerably accelerated the degradation process (k2 = 1.54093 L mg-1min-1). The material was applied in a synthetic effluent containing various ions, and its performance consistently approached 100% in the photo-Fenton system. Finally, two degradation byproducts were identified through HPLC-MS/MS analysis.

Modeling Recurrent Failures on Large Directed Networks

Many lifeline infrastructure systems consist of thousands of components configured in a complex directed network. Disruption of the infrastructure constitutes a recurrent failure process over a directed network. Statistical inference for such network recurrence data is challenging because of the large number of nodes with irregular connections among them. Motivated by 16 years of Scottish Water operation records, we propose a network Gamma-Poisson Autoregressive NHPP (GPAN) model for recurrent failure data from large-scale directed physical networks. The model consists of two layers: the temporal layer applies a Non-Homogeneous Poisson Process (NHPP) with node-specific frailties, and the spatial layer uses a well-orchestrated gamma-Poisson autoregressive scheme to establish correlations among the frailties. Under the network-GPAN model, we develop a sum-product algorithm to compute the marginal distribution for each frailty conditional on the recurrence data. The marginal conditional frailty distributions are useful for predicting future failures based on historical data. In addition, the ability to rapidly compute these marginal distributions allows adoption of an EM type algorithm for estimation. Through a Bethe approximation, the output from the sum-product algorithm is used to compute maximum log-likelihood estimates. Applying the methods to the Scottish Water network, we demonstrate utility in aiding operation management and risk assessment of the water utility. Supplementary materials for this article are available online including a standardized description of the materials available for reproducing the work.

Unsupervised Gene-Cell Collective Representation Learning with Optimal Transport

Cell type identification plays a vital role in single-cell RNA sequencing (scRNA-seq) data analysis. Although many deep embedded methods to cluster scRNA-seq data have been proposed, they still fail in elucidating the intrinsic properties of cells and genes. Here, we present a novel end-to-end deep graph clustering model for single-cell transcriptomics data based on unsupervised Gene-Cell Collective representation learning and Optimal Transport (scGCOT) which integrates both cell and gene correlations. Specifically, scGCOT learns the latent embedding of cells and genes simultaneously and reconstructs the cell graph, the gene graph, and the gene expression count matrix. A zero-inflated negative binomial (ZINB) model is estimated via the reconstructed count matrix to capture the essential properties of scRNA-seq data. By leveraging the optimal transport-based joint representation alignment, scGCOT learns the clustering process and the latent representations through a mutually supervised self optimization strategy. Extensive experiments with 14 competing methods on 15 real scRNA-seq datasets demonstrate the competitive edges of scGCOT.

Does digital finance contribute to regional economic growth? Evidence from RCEP countries

This study examines the developmental characteristics of the digital finance cooperation of China and the Regional Comprehensive Economic Partnership (RCEP) in the global economic transition to the digital era. Using a fixed-effects model of panel data from RCEP countries from 2011 to 2021, the study explores the mechanisms between digital finance development and regional economic growth. Furthermore, it conducts a spatial analysis to obtain geographic variables, using their interaction term with internet penetration as instrumental variables. The results show that digital finance development can effectively promote regional economic growth, and an endogeneity test confirms its robustness.

Abstract 908: SeNMo: A self-normalizing deep learning model for enhanced multi-omics data analysis in oncology

Abstract Multi-omics research has enhanced our understanding of cancer heterogeneity and progression. Investigating molecular data through multi-omics approaches is crucial for unraveling the complex biological mechanisms underlying cancer, thereby enabling more effective diagnosis, treatment, and prevention strategies. However, predicting patient outcomes through integration of all available multi-omics data is still an under-study research direction. Here, we present SeNMo (Self-normalizing Network for Multi-omics), a deep neural network that ensures the zero mean and unit variance of activations across network layers using the self-normalizing technique. Such normalizing techniques are critical in stable and robust learning of deep learning models. SeNMo is particularly efficient in handling multi-omics data characterized by high-width (many features) and low-length (fewer samples) attributes. We trained SeNMo for the task of overall survival of patients using pan-cancer multi-omics data involving 28 cancer sites from the Genomic Data Commons (GDC). The training multi-omics data includes gene expression, DNA methylation, miRNA expression, and protein expression modalities. We tested the model's performance on the Moffitt Cancer Center's internal data involving RNA expression and protein expression data. We evaluated the model’s performance in predicting patient’s overall survival using the concordance index (C-Index), which provides a robust measure of the model's predictive capability. SeNMo performed consistently well in the training regime, reflected by the validation C-Index≥0.6 on GDC's public data. In the testing regime on Moffitt's private data, SeNMo performed with a C-Index of 0.68. The model's performance increased when tested on low-dimensional data or when tested on single omic data such as RNA or protein expression data with a C-Index of 0.7. SeNMo proved to be a mini-foundation model for multi-omics oncology data because it demonstrated robust performance, adaptability across molecular data types, and universal approximator capabilities for the scale of molecular data it was trained on. SeNMo can be further scaled to any cancer site and molecular data type. It can also be fine-tuned for other downstream tasks such as treatment response prediction, risk stratification, patient subgroup identification, and others. Its ability to accurately predict patient outcomes and adapt to various downstream tasks indicates a new era in cancer research and treatment. For future research, SeNMo offers a powerful tool for uncovering deeper insights into the complex nature of cancer and sets a precedent for how artificial intelligence can be leveraged to handle the vast and intricate data in the biomedical field. We believe SeNMo and similar models are poised to transform the oncology landscape, offering hope for more effective, efficient, and patient-centric cancer care. Citation Format: Asim Waqas, Aakash Tripathi, Sabeen Ahmed, Ashwin Mukund, Paul Stewart, Mia Naeini, Hamza Farooq, Ghulam Rasool. SeNMo: A self-normalizing deep learning model for enhanced multi-omics data analysis in oncology [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 908.

Statistical post-processing of numerical weather prediction data using distribution-based scaling for wind energy

The performance of numerical weather prediction models has improved dramatically recently. However, model biases remain a fundamental limitation prohibiting the direct implementation of model results. There are several ways to describe wind speed data. The Weibull and lognormal distributions are used to obtain the best-fit model for the wind speed data. This study aims to develop a statistical post-processing method based on the distribution-based scaling (DBS) approach, which scales NWP data to fit the distribution derived using recorded wind speed at that site location. The performance of the suggested method was evaluated using four different error measures. The optimal model is anticipated to have the lowest Mean Bias Error (MBE), Mean Absolute Error (MAE), Root Mean square Error (RMSE), and variance (s2) values. It was determined that employing a DBS strategy significantly improved the NWP by at least 75%.

The Development of Forecasting Models for Life Insurance Data by Employing Time-series Analysis and Machine Learning Technique

This article is conducted with the primary objective of investigating and comparing various forecasting models, aiming to identify the optimal model for life insurance data. For this investigation, we have employed a comprehensive dataset containing monthly direct premium data from the Thai life insurance sector, spanning from January 2003 to December 2022. Our approach involves the development of time-series models to forecast direct premiums, initially employing the SARIMAX framework. Subsequently, we have introduced an additional time-series forecasting model that incorporates SVR, collectively referred to as the SVR-SARIMAX model. The evaluation criteria used for model comparison encompass the Mean Absolute Percentage Error (MAPE), Root Mean Square Error (RMSE), and the Coefficient of Determination (R2). The results of our analysis demonstrate that the SARIMAX model outperforms both the SVR and SVR-SARIMAX models, primarily due to the linear pattern in the relationship between the independent and dependent variables. Nevertheless, it is noteworthy that the proposed SVR-SARIMAX model exhibits an improvement in prediction accuracy compared to the standalone non-linear model (SVR), even though the linear model (SARIMAX) still demonstrates superior accuracy.

A note on the covariate-dependent kink threshold regression model for panel data

This article provides new estimating and testing procedures for the panel covariate-dependent kink threshold regression model. By utilizing a linearization technique, we develop an efficient estimator that offers a computationally viable alternative to the grid-search approach. Then we propose a novel testing procedure based on the score statistic to examine the existence of the covariate-dependent kink effect. The test statistic only requires fitting the model in the absence of varying kink threshold, greatly reducing computing time. Extensive simulation studies demonstrate that the proposed method outperforms existing methodologies. To further illustrate their performance, we utilize a real-data analysis. The results demonstrate the practicality and effectiveness of our approach in a wide range of scenarios.

Analyzing the 20-year declining trend of hospital length-of-stay in European countries with different healthcare systems and reimbursement models

The study aims to investigate the last 20-year (2000–2019) of hospital length of stay (LOS) trends and their association with different healthcare systems (HS) among 25 European countries. A panel dataset was created using secondary data from Eurostat and Global Burden of Disease study databases, with dependent and control variables aggregated at the national level over a period of 20 years. A time trend analysis was conducted using a weighted least squares model for panel data to investigate the association between LOS, HS models [National Health Service (NHS), National Health Insurance, Social Health Insurance (SHI), and Etatist Social Health Insurance], healthcare reimbursement schemes [Prospective Global Budget (PGB), Diagnosis Related Groups (DRG), and Procedure Service Payment (PSP)], and control variables. The study showed a reduction of average LOS from 9.20 days in 2000 to 7.24 in 2019. SHI was associated with a lower LOS compared to NHS (b = − 0.6327, p < 0.05). Both DRG (b = 1.2399, p < 0.05) and PSP (b = 1.1677, p < 0.05) reimbursement models were positively associated with LOS compared to PGB. Our results confirmed the downward trend of LOS in the last 20 years, its multifactorial nature, and the influence of the SHI model of HS. This could be due to the financial incentives present in fee-for-service payment models and the role of competition in creating a market for healthcare services. These results offer insight into the factors influencing healthcare utilization and can inform the design of more effective, efficient, and sustainable HS.

Prediction uncertainty estimates elucidate the limitation of current NSCLC subtype classification in representing mutational heterogeneity

The heterogeneous pathogenesis and treatment response of non-small cell lung cancer (NSCLC) has led clinical treatment decisions to be guided by NSCLC subtypes, with lung adenocarcinoma and lung squamous cell carcinoma being the most common subtypes. While histology-based subtyping remains challenging, NSCLC subtypes were found to be distinct at the transcriptomic level. However, unlike genomic alterations, gene expression is generally not assessed in clinical routine. Since subtyping of NSCLC has remained elusive using mutational data, we aimed at developing a neural network model that simultaneously learns from adenocarcinoma and squamous cell carcinoma samples of other tissue types and is regularized using a neural network model trained from gene expression data. While substructures of the expression-based manifold were captured in the mutation-based manifold, NSCLC classification accuracy did not significantly improve. However, performance was increased when rejecting inconclusive samples using an ensemble-based approach capturing prediction uncertainty. Importantly, SHAP analysis of misclassified samples identified co-occurring mutations indicative of both NSCLC subtypes, questioning the current NSCLC subtype classification to adequately represent inherent mutational heterogeneity. Since our model captures mutational patterns linked to clinical heterogeneity, we anticipate it to be suited as foundational model of genomic data for clinically relevant prognostic or predictive downstream tasks.

CNN Based Face Emotion Recognition System for Healthcare Application

INTRODUCTION: Because it has various benefits in areas such psychology, human-computer interaction, and marketing, the recognition of facial expressions has gained a lot of attention lately.&#x0D; OBJECTIVES: Convolutional neural networks (CNNs) have shown enormous potential for enhancing the accuracy of facial emotion identification systems. In this study, a CNN-based approach for recognizing facial expressions is provided. METHODS: To boost the model's generalizability, transfer learning and data augmentation procedures are applied. The recommended strategy defeated the existing state- of-the-art models when examined on multiple benchmark datasets, including the FER-2013, CK+, and JAFFE databases. &#x0D; RESULTS: The results suggest that the CNN-based approach is fairly excellent at properly recognizing face emotions and has a lot of potential for usage in detecting facial emotions in practical scenarios.&#x0D; CONCLUSION: Several diverse forms of information, including oral, textual, and visual, maybe applied to comprehend emotions. In order to increase prediction accuracy and decrease loss, this research recommended a deep CNN model for emotion prediction from facial expression.

Effectiveness of a Digital Health Game Intervention on Early Adolescent Smoking Refusal Self-Efficacy.

Smoking poses a significant threat to adolescent health because of its immediate and long-term detrimental health effects. Smoking refusal self-efficacy predicts smoking behavior in adolescence. In adolescents' health education, digital interventions are potential tools to support smoking refusal self-efficacy. The aim of this two-arm cluster randomized controlled trial was to evaluate the effectiveness of a digital health game intervention compared with a no-intervention control group on smoking refusal self-efficacy in 10- to 13-year-old Finnish early adolescents. The early adolescents (n = 781) were randomized to the control group (n = 394) and the health game intervention group (n = 387). Smoking refusal self-efficacy, sources of smoking and snus refusal self-efficacy, and motivation to decline smoking and snus use in the future were measured at baseline, 2-week postintervention, and 3-month follow-up. Data were analyzed using linear mixed model and Wilcoxon rank-based test for clustered data. According to the results, the intervention group made improvements in sources of smoking and snus refusal self-efficacy between baseline and postintervention, and in sources of snus refusal self-efficacy between baseline and follow-up, compared with the control group. The intervention group showed improvements in smoking refusal self-efficacy among 12-year-olds between baseline and follow-up, and postintervention and follow-up compared with the control group. Similar improvements were also found among those with a smoking friend or a smoking parent between postintervention and follow-up. The results were promising for the use of digital health game interventions to promote early adolescent smoking refusal self-efficacy and preventing smoking experimentation. Further research can evaluate the long-term effects for adolescents.