Method For Models Research Articles

A Fractal Adsorption Model on Methane in Coal with Temperature Effect Dependence

The traditional Langmuir equation displays drawback in accurately characterizing the methane adsorption behavior in coal, due to it assuming the uniform surface of coal pores. Additionally, the decay law of gas adsorption capacity with an increasing coal reservoir temperature remains unknown. In this study, the fractal adsorption model is proposed based on the fractal dimension (Df) of coal pores and the attenuation coefficient (n) of the adsorption capacity. The principles and methods of this fractal adsorption model are deduced and summarized in detail. The results show that the pore structures of the two coal samples exhibit obvious fractal characteristics, with the values of fractal dimensions (Df) being 2.6279 and 2.93. The values of adsorption capacity attenuation coefficients (n) are estimated as −0.006 and −0.004 by the adsorption experiments with different temperatures. The proposed fractal adsorption model presents a greater theoretical significance and higher accuracy than that of the Langmuir equation. The accuracy of the fractal adsorption model with temperature effect dependence is verified, establishing a prediction method for methane adsorption capacity in deep coal reservoirs. This study can serve as a theoretical foundation for coalbed methane exploration and development, as well as provide valuable insights for unconventional natural gas exploitation.

Site selection and prediction of urban emergency shelter based on VGAE-RF model

As urban development accelerates and natural disasters occur more frequently, the urgency of developing effective emergency shelter planning strategies intensifies. The shelter location selection method under the traditional multi-criteria decision-making framework suffers from issues such as strong subjectivity and insufficient data support. Artificial intelligence offers a robust data-driven approach for site selection; however, many methods neglect the spatial relationships of site selection targets within geographical space. This paper introduces an emergency shelter site selection model that combines a variational graph autoencoder (VGAE) with a random forest (RF), namely VGAE-RF. In the constructed urban spatial topological graph, based on network geographic information, this model captures both the latent features of geographic unit coupling and integrates explicit and latent features to forecast the likelihood of emergency shelters in the construction area. This study takes Beijing, China, as the experimental area and evaluates the reliability of different model methods using a confusion matrix, Receiver Operating Characteristic (ROC) curve, and Imbalance Index of spatial distribution as evaluation indicators. The experimental results indicate that the proposed VGAE-RF model method, which considers spatial semantic associations, displays the best reliability.

Estimation of Standard Error, Linking Error, and Total Error for Robust and Nonrobust Linking Methods in the Two-Parameter Logistic Model

The two-parameter logistic (2PL) item response theory model is a statistical model for analyzing multivariate binary data. In this article, two groups are brought onto a common metric using the 2PL model using linking methods. The linking methods of mean–mean linking, mean–geometric–mean linking, and Haebara linking are investigated in nonrobust and robust specifications in the presence of differential item functioning (DIF). M-estimation theory is applied to derive linking errors for the studied linking methods. However, estimated linking errors are prone to sampling error in estimated item parameters, thus resulting in artificially increased the linking error estimates in finite samples. For this reason, a bias-corrected linking error estimate is proposed. The usefulness of the modified linking error estimate is demonstrated in a simulation study. It is shown that a simultaneous assessment of the standard error and linking error in a total error must be conducted to obtain valid statistical inference. In the computation of the total error, using the bias-corrected linking error estimate instead of the usually employed linking error provides more accurate coverage rates.

Risk Groups and Psychosocial Factors for the Pandemic (COVID-19).

COVID-19, which started in 2019 and affected the whole world, has affected everyone at different intensities and in different ways. COVID-19, which is considered a pandemic, has turned into a major public health problem in terms of its consequences and has affected people biopsychosocially. However, people in risk groups may be affected more. This study was conducted to reveal the risk groups for the pandemic and to determine the psychosocial factors. Data were collected online using the relational screening model and snowball sampling methods. A Sociodemographic Information Form, COVID-19 Pandemic Psychosocial Impact Scale (C19-PPIS), and International Personality Inventory Short Form (IPISV) were sent online to 826 participants. Data were analysed using an independent sample t-test, a one-way ANOVA test, and the Pearson Correlation analysis. According to the results, young adults (X¯ = 2.77), women (X¯ = 2.79), singles (X¯ = 2.78), those who are unemployed (X¯ = 2.89), and those who had to change their home or city due to the pandemic (X¯ = 2.89) were more affected by the pandemic. Psychological support was the support system needed the most during the pandemic (X¯ = 3.04). In addition, a negative relationship was found between an extroverted personality and psychosocial impact from the pandemic (r = -0.148 and p < 0.01). A positive relationship was found between introversion (r = 0.183 and p < 0.01), agreeableness (r = 0.078 and p < 0.05), hostility (r = 0.094 and p < 0.01), disorganisation (r = 0.237 and p < 0.01), openness to development (r = 0.80 and p < 0.05), closed off to development (r = 0.070 and p < 0.05), emotional instability personality (r = 0.498 and p < 0.01), and psychosocial impact from the pandemic. This study has revealed important results regarding who has been most affected psychosocially by COVID-19. It is thought that the results obtained can guide state policies on what should be done in the field of preventive community mental health in another possible epidemic.

Dynamic characteristics of vertically irregular structures with random fields of different probability distributions based on stochastic homotopy method

Considering the uncertainty of elastic modulus in different materials as well as the strong nonlinearity between the structural frequency and elastic modulus, it is a great challenge for computing the dynamic characteristics of vertically irregular structures. In this research, a new stochastic homotopy approach is presented to compute the basic dynamic characteristics of the vertically irregular structures, where the elastic modulus of different materials are assumed as the random fields with different probability distributions. To obtain the dynamic characteristics of the vertically irregular structures, a stochastic eigenvalue equation is established firstly. Then the stochastic eigenvalue and eigenvector are represented by the homotopy series, and the stochastic residual error in relation to the stochastic eigenvalue equation is minimized to obtain the coefficients of the homotopy series. Afterwards, the basic dynamic characteristics, including stochastic natural frequencies and modal shapes, of the vertically irregular structures are obtained. In comparison to the sampling-based stochastic surrogate model methods like the Kriging and non-intrusive arbitrary polynomial chaos methods, the presented approach can efficiently yield more stable statistical moments of the natural frequencies. Meanwhile the proposed method can generate the statistical moments of the modal shapes, which is difficult to achieve with the stochastic surrogate model methods. Finally, the effectiveness of the suggested method is testified through two examples of a reinforced concrete-steel column and a realistic reinforced concrete-steel frame structure.

Integrated control strategy for 5G base station frequency regulation considering spatio-temporal variations in communication load

The decreasing system inertia and active power reserves caused by the penetration of renewable energy sources and the displacement of conventional generating units present new challenges to the frequency stability of modern power systems. Vast quantities of 5G base stations, featuring largely dormant battery storage systems and advanced communication technology, represent a high-quality fast frequency regulation resource for the power system. This paper proposes a double-layer clustering method for 5G base stations and an integrated centralized-decentralized control strategy for their participation in frequency regulation, considering the variability of available frequency regulation capacity due to spatial and temporal variations in base station communication loads. The proposed capacity model and control methods are evaluated using a case study of a two-machine test system with 10,000 real 5G base stations, demonstrating the effectiveness of the clustering method and the integrated control strategy in improving frequency stability.

Trends, turning points, and driving forces of desertification in global arid land based on the segmental trend method and SHAP model

ABSTRACT Desertification is a form of land degradation observed in arid, semiarid, and dry subhumid ecosystems. Assessing the global trends and drivers of desertification in arid land is crucial for developing effective land restoration policies and mitigating desertification. We aimed to evaluate global segmental trends in desertification in arid land using Moderate Resolution Imaging Spectroradiometer images from 2000 to 2022. By constructing a robust MSAVI-Albedo Desertification Distance Index (DDI), we assessed the segmented development characteristics of desertification on Google Earth Engine. Additionally, we employed the SHapley Additive exPlanations (SHAP) model and machine learning methods to analyze the individual and interactive driving mechanisms of desertification. The results indicated an overall reduction in desertification, with approximately 23.21 million km2 (39% of the global arid region area) exhibiting negative trends in DDI. Approximately 31% of the land area showed a DDI of −0.002/y. Precipitation was consistently the primary factor influencing desertification, with an average SHAP value of 11.42. Secondary influencing factors included potential evapotranspiration, soil moisture, and vapor pressure differences. Notably, the coupling between precipitation and soil moisture exhibited the most significant impact on the desertification process, with SHAP coupling values of approximately 3.28 and 5.06 before and after the DDI turning point, respectively. These findings provide new insights into desertification on a global scale and offer valuable scientific support for promoting effective prevention and control measures.

Causal Associations of Environmental Pollution and Cardiovascular Disease: A Two-Sample Mendelian Randomization Study.

There is growing evidence that concentrations of environmental pollutants are previously associated with cardiovascular disease; however, it is unclear whether this association reflects a causal relationship. We utilized a two-sample Mendelian randomization (MR) approach to investigate how environmental pollution affects the likelihood of developing cardiovascular disease. We primarily employed the inverse variance weighted (IVW) method. Additionally, to ensure the robustness of our findings, we conducted several sensitivity analyses using alternative methodologies. These included maximum likelihood, MR-Egger regression, weighted median method and weighted model methods. Inverse variance weighted estimates suggested that an SD increase in PM2.5 exposure increased the risk of heart failure (OR = 1.40, 95% CI 1.02-1.93, p = 0.0386). We found that an SD increase in PM10 exposure increased the risk of hypertension (OR = 1.45, 95% CI 1.02-2.05, p = 0.03598) and atrial fibrillation (OR = 1.41, 95% CI 1.03-1.94, p = 0.03461). Exposure to chemical or other fumes in a workplace was found to increase the risk of hypertension (OR = 3.08, 95% CI 1.40-6.78, p = 0.005218), coronary artery disease (OR = 1.81, 95% CI 1.00-3.26, p = 0.04861), coronary heart disease (OR = 3.15, 95% CI 1.21-8.16, p = 0.0183) and myocardial infarction (OR = 3.03, 95% CI 1.13-8.17, p = 0.02802). This study reveals the causal relationship between air pollutants and cardiovascular diseases, providing new insights into the protection of cardiovascular diseases.

Research on a Human Moving Object Detection Method Based on Gaussian Model and Deep Learning

In order to understand human motion object detection methods, the author proposes a research on human motion object detection method based on Gaussian model. Firstly, traditional Gaussian models are unable to detect complex scenes or slow moving targets. Therefore, an improved Gaussian model based moving object detection algorithm is proposed. Secondly, multiple Gaussian models are used to represent the features of each pixel in the moving target image, and based on the matching of each pixel in the image with the Gaussian model, it is considered as a background point. Conversely, it is based on the principle of the foreground, and the Gaussian model is updated. Finally, by updating the foreground model and calculating short-term stability indicators, the detection effect of moving targets is improved. By determining the Gaussian distribution and pixel relationship, new parameters are set to construct the background model and eliminate the impact caused by sudden changes in lighting. The experimental analysis results show that this method can effectively detect and track moving targets, with good noise resistance, high clarity, and an accuracy rate of up to 99%. Compared with traditional Gaussian model methods, the improved method can more effectively detect moving targets and has better robustness.

Unveiling the relation between digital technology and low-carbon innovation: Carbon emission trading policy as an antecedent

While the carbon emission trading policy presents an opportunity for firms to concurrently advance in digital technology and low-carbon innovation, the existing literature falls short in elucidating the relationship between the two. Therefore, based on the sample of Chinese A-share listed companies from 2007 to 2021, this study examines the impact of digital technology on low-carbon innovation under the carbon emission trading policy by using DID model, moderating variable and other methods. The results show that digital technology is positively correlated with low-carbon innovation under the carbon emission trading policy. Secondly, digital technology breadth and depth emerge as dual facilitative avenues. However, carbon emission trading schemes appears to curb the impact of the digital technology depth. Thirdly, technological capability is a key resource in moderating the relationship between digital technology and low-carbon innovation. Internal control, however, exerts a significant moderating effect exclusively on digital technology breadth. Finally, drawing from the analysis of company differentiation, the study conclude that smaller firms and research and development (R&D)-oriented firms are better poised to implement strategic plans that harmonize digital technology and low-carbon innovation.

Prominent User Segments in Online Consumer Recommendation Communities: Capturing Behavioral and Linguistic Qualities with User Comment Embeddings

Online conversation communities have become an influential source of consumer recommendations in recent years. We propose a set of meaningful user segments which emerge from user embedding representations, based exclusively on comments’ text input. Data were collected from three popular recommendation communities in Reddit, covering the domains of book and movie suggestions. We utilized two neural language model methods to produce user embeddings, namely Doc2Vec and Sentence-BERT. Embedding interpretation issues were addressed by examining latent factors’ associations with behavioral, sentiment, and linguistic variables, acquired using the VADER, LIWC, and LFTK libraries in Python. User clusters were identified, having different levels of engagement and linguistic characteristics. The latent features of both approaches were strongly correlated with several user behavioral and linguistic indicators. Both approaches managed to capture significant variability in writing styles and quality, such as length, readability, use of function words, and complexity. However, the Doc2Vec features better described users by varying level of contribution, while S-BERT-based features were more closely adapted to users’ varying emotional engagement. Prominent segments revealed prolific users with formal, intuitive, emotionally distant, and highly analytical styles, as well as users who were less elaborate, less consistent, but more emotionally connected. The observed patterns were largely similar across communities.

The Quest for an ESG Country Rank: A Performance Contribution Analysis/MCDM Approach

Utilizing Multi-Criteria Decision Analysis (MCDA) methods based on environmental, social, and governance (ESG) factors to rank countries according to these criteria aims to evaluate and prioritize countries based on their performance in environmental, social, and governance aspects. The contemporary world is influenced by a multitude of factors, which consequently impact our lives. Various models are devised to assess company performance, with the intention of enhancing quality of life. An exemplary case is the ESG framework, encompassing environmental, social, and governmental dimensions. Implementing this framework is intricate, and many nations are keen on understanding their global ranking and avenues for enhancement. Different statistical and mathematical methods have been employed to represent these rankings. This research endeavors to examine both types of methods to ascertain the one yielding the optimal outcome. The ESG model comprises eleven factors, each contributing to its efficacy. We employ the Performance Contribution Analysis (PCA), Clifford algebra method, and entropy weight technique to rank these factors, aiming to identify the most influential factor in countries’ ESG-based rankings. Based on prioritization results, political stability (PSAV) and the voice of accountability (VA) emerge as pivotal elements. In light of the ESG model and MCDA methods, the following countries exhibit significant societal impact: Sweden, Finland, New Zealand, Luxembourg, Switzerland, Denmark, India, Norway, Canada, Germany, Austria, and Australia. This research contributes in two distinct dimensions, considering the global context and MCDA methods employed. Undoubtedly, a research gap is identified, necessitating the development of a novel model for the comparative evaluation of countries in relation to prior studies.

Contrastive pre-training for sequence based genomics models.

In recent years deep learning has become one of the central approaches in a number of applications, including many tasks in genomics. However, as models grow in depth and complexity, they either require more data or a strategic initialization technique to improve performance. In this project, we introduce cGen, a novel unsupervised, model-agnostic contrastive pre-training method for sequence-based models. cGen can be used before training to initialize weights, reducing the size of the dataset needed. It works through learning the intrinsic features of the reference genome and makes no assumptions on the underlying structure. We show that the embeddings produced by the unsupervised model are already informative for gene expression prediction and that the sequence features provide a meaningful clustering. We demonstrate that cGen improves model performance in various sequence-based deep learning applications, such as chromatin profiling prediction and gene expression. Our findings suggest that using cGen, particularly in areas constrained by data availability, could improve the performance of deep learning genomic models without the need to modify the model architecture.

Sustainability reporting and investor sentiment. A sustainable development approach to Chinese-listed firms.

This study highlights the influence of sustainability reporting on investor sentiments in the China Stock Exchange. The study starts by utilizing an Ordinary Least Squares regression model to test the hypotheses. Advanced econometric techniques are then applied to identify the existence of heteroskedasticity. To address potential endogeneity concerns, the analysis incorporates fixed-effect, two-stage least squares, and two-step generalized method of moments regression models. Findings suggest that sustainability reporting has a positive influence on investor sentiments. Conversely, environmental, social, and governance sustainability reporting also positively associations with investor sentiment in fixed-effect, two-stage least squares, and two-step generalized method of moments results. The findings suggest that companies prioritizing transparent and responsible practices enhance their market standing and contribute significantly to sustainable and ethical investing. The research indicates the importance of context-specific sustainability reporting. It provides insights into sustainability's impact on investor sentiments, promoting responsible practices for a sustainable global economy.

Flexi-BOPI: Flexible granularity pipeline inference with Bayesian optimization for deep learning models on HMPSoC

To achieve high-throughput deep learning (DL) model inference on heterogeneous multiprocessor systems-on-chip (HMPSoC) platforms, the use of pipelining for the simultaneous utilization of multiple resources has emerged as a promising solution. Nevertheless, current research faces two primary challenges: determining the optimal pipeline partitioning granularity, which directly influences the inference performance, and addressing the high time overhead of the search algorithms. To address these challenges, we propose Flexi-BOPI, a pipeline inference method for DL models of HMPSoCs. Flexi-BOPI offers flexible pipeline partitioning granularity down to a minimum size of a single core, enhancing the performance by better adapting to the diverse computational demands of different layers in DL models. Flexi-BOPI employs a Bayesian optimization-based search algorithm to significantly reduce the search overhead. In addition, we propose a surrogate model based on the heteroscedastic Gaussian process (HGP) to address the challenge of sample noise during the evaluation process. This approach can further reduce search overhead. Our experimental results demonstrate that the proposed method achieves significant improvements in inference performance and search overhead compared to existing methods.

TPoison: Data-Poisoning Attack against GNN-Based Social Trust Model

In online social networks, users can vote on different trust levels for each other to indicate how much they trust their friends. Researchers have improved their ability to predict social trust relationships through a variety of methods, one of which is the graph neural network (GNN) method, but they have also brought the vulnerability of the GNN method into the social trust network model. We propose a data-poisoning attack method for GNN-based social trust models based on the characteristics of social trust networks. We used a two-sample test for power-law distributions of discrete data to avoid changes in the dataset being detected and used an enhanced surrogate model to generate poisoned samples. We further tested the effectiveness of our approach on three real-world datasets and compared it with two other methods. The experimental results using three datasets show that our method can effectively avoid detection. We also used three metrics to illustrate the effectiveness of our attack, and the experimental results show that our attack stayed ahead of the other two methods in all three datasets. In terms of one of our metrics, our attack method decreased the accuracies of the attacked models by 12.6%, 22.8%, and 13.8%.

Medical Record Document Search with TF-IDF and Vector Space Model (VSM)

The growth of medical record documents is increasing over time, and the various types of diseases and therapies needed are increasing. However, this has not been followed by an effective and efficient search process. This study aims to deal with search problems that often take a long time with search results that are not necessarily as expected by building a search model for medical record documents using the vector space model (VSM) and TF-IDF methods. The VSM method allows retrieval of results that are not the same as the search queries entered by the user but are expected to provide still results relevant to the user's desired needs. The model development process was taken based on the data in the FS_ANAMNESA and FS_DIAGNOSA columns, followed by preprocessing, which consists of deleting blank lines, lowercase, removing punctuation marks, HTML tags, stop words, excess spaces between words, and normalizing typo words, then forming a TF-IDF matrix based on the frequency of occurrence of each word feature, and followed by the calculation of the similarity value of the search query compared to medical record documents based on the cosine similarity formula. The retrieval results were all columns of each existing medical record document and were sorted based on 10 rows with the highest similarity value. The model evaluation results were based on 1000 medical record documents and tested with 20 search queries in this study, which gave an average precision value of 0.548 and an average recall value of 0.796.

Improved Latin hypercube sampling initialization-based whale optimization algorithm for COVID-19 X-ray multi-threshold image segmentation

Image segmentation techniques play a vital role in aiding COVID-19 diagnosis. Multi-threshold image segmentation methods are favored for their computational simplicity and operational efficiency. Existing threshold selection techniques in multi-threshold image segmentation, such as Kapur based on exhaustive enumeration, often hamper efficiency and accuracy. The whale optimization algorithm (WOA) has shown promise in addressing this challenge, but issues persist, including poor stability, low efficiency, and accuracy in COVID-19 threshold image segmentation. To tackle these issues, we introduce a Latin hypercube sampling initialization-based multi-strategy enhanced WOA (CAGWOA). It incorporates a COS sampling initialization strategy (COSI), an adaptive global search approach (GS), and an all-dimensional neighborhood mechanism (ADN). COSI leverages probability density functions created from Latin hypercube sampling, ensuring even solution space coverage to improve the stability of the segmentation model. GS widens the exploration scope to combat stagnation during iterations and improve segmentation efficiency. ADN refines convergence accuracy around optimal individuals to improve segmentation accuracy. CAGWOA's performance is validated through experiments on various benchmark function test sets. Furthermore, we apply CAGWOA alongside similar methods in a multi-threshold image segmentation model for comparative experiments on lung X-ray images of infected patients. The results demonstrate CAGWOA's superiority, including better image detail preservation, clear segmentation boundaries, and adaptability across different threshold levels.

The effectiveness of natural reserves from the perspective of habitat quality in the southern section of the Hengduan Mountains, Southwestern China

Natural reserves (NRs) play key roles in habitat integrity conservation and biodiversity loss mitigation, and the assessment of the conservation effectiveness of NRs is needed to better manage them. Habitat quality (HQ) comprehensively reflects habitat integrity and biodiversity, but the conservation effectiveness of NRs from the perspective of HQ has rarely been determined at high spatial resolution. Taking the southern section of the Hengduan Mountains (SSHM) in Southwest China as an example, combining an InVEST-HQ model and spatiotemporal change detection methods, the effectiveness of NRs from the perspective of HQ at 30-m spatial resolution was assessed in this study. The effectiveness disparities of NRs across different properties (i.e., management level, conservation target, size, and establishment age) was analyzed and the human pressures on NRs was investigated. The results showed that the HQ of the NRs is good in the SSHM, with the area ratio of the Higher and Highest HQ ≥ 93%. Most of the NR area (94.11%) was effective at improving or maintaining a good HQ. With regard to NR properties, county NRs, NRs designated to conserving wild animals, middle NRs, and younger NRs were more effective, corresponding to management level, conservation target, size, and establishment age, respectively. The human footprint for an effective area is significantly lower than that for an ineffective area, consistent with higher HQ in the effective area and lower HQ in the ineffective area. These findings support the management and zoning of NRs in the SSHM to ensure their effectiveness.

Methodology for Severe Convective Cloud Identification Using Lightweight Neural Network Model Ensembling

This study introduces an advanced ensemble methodology employing lightweight neural network models for identifying severe convective clouds from FY-4B geostationary meteorological satellite imagery. We have constructed a FY-4B based severe convective cloud dataset by a combination of algorithms and expert judgment. Through the ablation study of a model ensembling combination of multiple specialized lightweight architectures—ENet, ESPNet, Fast-SCNN, ICNet, and MobileNetV2—the optimal EFNet (ENet- and Fast-SCNN-based network) not only achieves real-time processing capabilities but also ensures high accuracy in severe weather detection. EFNet consistently outperformed traditional, heavier models across several key performance indicators: achieving an accuracy of 0.9941, precision of 0.9391, recall of 0.9201, F1 score of 0.9295, and computing time of 18.65 s over the test dataset of 300 images (~0.06 s per 512 × 512 pic). ENet shows high precision but misses subtle clouds, while Fast-SCNN has high sensitivity but lower precision, leading to misclassifications. EFNet’s ensemble approach balances these traits, enhancing overall predictive accuracy. The ensemble method of lightweight models effectively aggregates the diverse strengths of the individual models, optimizing both speed and predictive performance.