Preferred Model Research Articles

Test Takers’ Attitudes Toward Varieties of Accents in Listening Tasks of the Duolingo English Test (2021 test version)

ABSTRACT There has been much debate in assessment research about the inclusion of Global English accents in high-stakes listening tests. This study explored test-takers’ attitudes toward the inclusion of different English accents in the Duolingo English Test (DET) 2021 test version and their associations with listening test scores. One hundred sixty English learners from four language backgrounds (Chinese, Korean, Hindi, and Latin American Spanish) completed yes/no vocabulary and dictation tasks that simulated the listening sections of the DET. The tasks included speech produced by English speakers of the same language background as the listeners, as well as American and British English. Learners completed a survey that elicited their attitudes toward non-standard English accents in proficiency tests. Exploratory factor analysis of survey responses revealed two contrasting trends in learners’ attitudes. Constructed responses suggested that while listeners generally preferred prestigious English models (e.g. American English or British English), they also expressed a need for incorporating other accent varieties. The relationships between listeners’ attitudes and their performance on the test were minimal (r < .26). The findings hint at a deeper understanding of test takers’ needs regarding accent varieties in listening tests. The study offers implications for the development of high-stakes English listening tests in global contexts.

Transition to Survivorship Care for Adolescents and Young Adults (AYAs) with Acute Leukemia: Provider Perspectives.

Purpose: Adolescent and young adult (AYA) survivors of acute lymphoblastic or myeloid leukemia diagnosed between the ages of 15 and 39 years are at risk for adverse late health effects following cancer treatment and require ongoing survivorship care. This study aims to understand the landscape of transitioning AYAs with leukemia from active treatment to survivorship care. Methods: A cross-sectional, anonymous online survey was sent out via listserv/email. Descriptive analyses were used to identify the proportion of providers whose institutions have AYA and/or survivorship programs, current practices in transitioning care, perspective on barriers and facilitators to transition, and preferred models of survivorship care. Differences by provider (i.e., adult medical oncologist vs. pediatric) were analyzed using chi-square or analysis of variance (ANOVA) analyses. Results: A total of 75 provider responses were analyzed; 51.4% of providers reported their institutions had an AYA program and over 80% had a survivorship program. Providers preferred the primary oncology team or survivorship specialist to address most survivorship care needs, except for the treatment of other comorbid conditions (primary care provider [PCP] preferred). Disease-related risks, lack of a PCP, and insurance concerns were commonly endorsed barriers to transition of care. Medical oncologists were more likely than pediatric providers to transition surveillance of new malignancies to PCPs. Conclusions: Providers preferred survivorship specialists to oversee AYA survivorship care, yet several barriers and concerns remain to implement the transition. Results indicate variability in transition care practices and preferences, particularly between medical and pediatric oncologists. Future work is needed to optimize transition practices to improve care for AYA survivors.

Prostaglandin D2 delays CD8+ T-cell responses and respiratory syncytial virus clearance in geriatric cotton rats.

Respiratory syncytial virus (RSV) infection is associated with increased rates of severe disease, hospitalization, and death in elderly individuals. Clearance of RSV is frequently delayed within this demographic, contributing to the more severe disease course. Geriatric cotton rats mimic this prolonged clearance kinetic and serve as a useful animal model for studying age-associated immunological deficits during RSV infection. Treatment with the cyclooxygenase (COX) inhibitor ibuprofen restores RSV clearance, indicating that inflammation contributes to impaired clearance in geriatric cotton rats. Here, we further characterize a compromised immune response in geriatric cotton rats and identify an inflammatory pathway that contributes to this deficiency. Dendritic cell (DC) activation and migration to mediastinal lymph nodes are decreased during early infection in geriatric cotton rats, resulting in delayed generation of cytotoxic T cells and virus clearance. Prostaglandin D2 (PGD2), which reduces DC migration through the elevation of D-type prostanoid 1 receptor (DP1 receptor), is elevated in the airways of infected geriatric cotton rats. Reducing PGD2 production by inhibiting COX-2 or PGD2 synthase improves RSV clearance kinetics through DC activation and RSV-specific CD8+ T-cell responses in geriatric cotton rats, whereas activation of DP1 receptor through an agonist resulted in delayed viral clearance in adult cotton rats. These results indicate that PGD2 contributes to delayed antigen presentation and CD8+ T-cell responses to RSV in geriatric cotton rats. Inhibiting PGD2 generation or signaling may be a useful mechanism of therapeutic intervention in elderly individuals.IMPORTANCEElderly adults are at increased risk of severe disease resulting from infection with respiratory syncytial virus (RSV), characterized in part by delayed clearance (removal of the virus from airways). Understanding the immunological factors that lead to this delayed clearance may allow for the development of therapies to improve disease outcomes in elderly individuals infected with RSV and other respiratory viruses. Here, we describe an inflammatory pathway in geriatric cotton rats, the preferred small animal laboratory model for RSV, that impairs the generation of an effective immune response. We show that inhibiting this inflammatory pathway in geriatric cotton rats improves immune parameters and speeds clearance of RSV. These results contribute to our understanding of delayed RSV clearance in elderly individuals with possible applications for improving immune responses to RSV in clinical settings.

Impact of Local Extent of Tumor on the Survival Outcomes for Surgically Treated Tongue Cancers.

The current classification of tongue cancers does not discriminate stages based on factors of local spread. Integrating factors of local spread that impact disease-specific survival (DSS) in a modified classification to improve prognostication compared with the current staging. This was a retrospective analysis of 399 previously untreated oral tongue squamous carcinomas operated between 2016 and 2018. Factors related to the local extent of a primary tumor that impacted the DSS were studied. Multivariable analyses adjusted for pT, pN stage, and adjuvant therapy. Candidate staging systems were developed based on factors that significantly impacted DSS, and validated. Base tongue involvement (p = 0.021), DOI > 20 mm (p value = 0.040), and tumor crossing the midline (p value = 0.047) significantly impacted the DSS. The preferred model included base tongue involvement in T3 stage, and DOI > 20 mm and tumor crossing the midline into T4 stage (Model 2), which was superior to the current classification system in AIC (830.55 vs. 845.43), C-index (0.65 vs. 0.61), and visual inspection of Kaplan-Meier curves. Identifying and possibly including these clinical factors in addition to the current T-stage criteria could result in better prognostication of tongue cancers. Integration of proposed classification could improve the discrimination between stages.

Preference learning based deep reinforcement learning for flexible job shop scheduling problem

The flexible job shop scheduling problem (FJSP) holds significant importance in both theoretical research and practical applications. Given the complexity and diversity of FJSP, improving the generalization and quality of scheduling methods has become a hot topic of interest in both industry and academia. To address this, this paper proposes a Preference-Based Mask-PPO (PBMP) algorithm, which leverages the strengths of preference learning and invalid action masking to optimize FJSP solutions. First, a reward predictor based on preference learning is designed to model reward prediction by comparing random fragments, eliminating the need for complex reward function design. Second, a novel intelligent switching mechanism is introduced, where proximal policy optimization (PPO) is employed to enhance exploration during sampling, and masked proximal policy optimization (Mask-PPO) refines the action space during training, significantly improving efficiency and solution quality. Furthermore, the Pearson correlation coefficient (PCC) is used to evaluate the performance of the preference model. Finally, comparative experiments on FJSP benchmark instances of varying sizes demonstrate that PBMP outperforms traditional scheduling strategies such as dispatching rules, OR-Tools, and other deep reinforcement learning (DRL) algorithms, achieving superior scheduling policies and faster convergence. Even with increasing instance sizes, preference learning proves to be an effective reward mechanism in reinforcement learning for FJSP. The ablation study further highlights the advantages of each key component in the PBMP algorithm across performance metrics.

Computer-aided design and fabrication of statistically shaped nasal prostheses: A feasibility study.

A nasal prosthesis may compensate for a partial or complete defect of the nose associated with trauma or amputation. However, the design and production is time-consuming, expensive, and expertize-dependent. Computer-generated prosthesis models and 3D printing can optimize the process. The purpose of this pilot study was to evaluate a novel method for nasal prosthesis design utilizing statistical shape modeling (SSM) on a cohort of participants. The prosthesis shapes were generated by a design algorithm and fabricated through a 3-dimensionally (3D) printed mold. Combining a semi-automated algorithm with 3D printing reduced the dependence on expertize. A proof-of-concept assessment was conducted by enrolling 20 participants with total nose amputation with a conventionally created prosthesis (Cp). The faces of the participants were scanned with an optical scanner to acquire 3D surface scans that were loaded into the algorithm to create the statistically shape modeled prosthesis (SSMp). Participants were asked to digitally modify the SSMp by changing 9 parameters using the Scalismo Lab software program, allowing them to shape their preferred nose model, the patient-preferred prosthesis (PPp). The SSMps were fabricated using 3D printed molds and mailed to the participants for testing. They were asked to complete questionnaires comprising three FACE-Q modules: satisfaction with the nose and psychological and social function. In addition, the SSMp and Cp were compared with the PPp using anthropometric measurements. Fifteen participants were able to fit the SSMp and complete the questionnaire. The median satisfaction, psychological, and social function scores were 56, 67, and 62, respectively. All 20 participants filled out the questionnaires on their Cp, with scores of 71, 67, and 62, respectively. Anthropometric measurements comparing the SSMp with the PPp showed that the algorithm generates noses with a wider nasal root, a larger nostril floor width to nasal width ratio, a smaller columella length to nasal width as well as nostril floor width ratio than the patients prefer according to their set PPp. Comparing the Cp with the PPp showed a larger nasal width to nasal height ratio and nasal root width to intercanthal distance ratio. SSM can facilitate the design and fabrication of nasal prostheses. Using the current models, however, patients were less satisfied than with prostheses made by the anaplastologist in consultation with the patient. Improving the SSM algorithm and incorporating patient manipulations into the resulting model should increase patient satisfaction and enable the production of nasal prostheses with minimal expert dependence.

An Evaluation of Rare Cancer Policies in Europe: A Survey Among Healthcare Providers.

Rare cancers, defined as those with an annual incidence of fewer than six cases per 100,000 individuals, are associated with significant health inequalities. This study aimed to assess the knowledge, attitudes, and perceptions of healthcare providers with expertise in rare cancers regarding the effectiveness of enacted or planned rare cancer policies across Europe. Between 25 March 2023 and 5 March 2024, we conducted an online survey targeting 738 healthcare providers affiliated with the European Reference Networks and the Organization of European Cancer Institutes, yielding 92 complete responses from 28 European countries (response rate: 12.5%). While a significant portion of respondents were unaware of their country's legal definitions for rare cancers, 67.4% acknowledged that national cancer plans prioritized rare cancers. These plans received the highest ratings for their evidence-based interventions and monitoring efforts. The integration of rare cancer policies into broader oncology frameworks was the preferred policy model. National cancer registries were highly rated for confidentiality and validity but scored the lowest for cost-effectiveness. Government funding was deemed crucial for cancer screening programs. The disease burden and unmet health needs primarily influenced reimbursement decisions in the field of rare cancers. Respondents rated palliative care as more effective in adults with rare cancers compared to in children, particularly regarding symptom management. We confirmed significant variability in rare cancer policy evaluations across Europe, the necessity for a common EU-level definition for rare cancers, and a shift in reimbursement and policy framework models, highlighting the importance of policy integration and enhanced collaboration. However, given the limitations of the study, such as small sample size and possible unstudied confounding factors, we should interpret our findings with caution. A systematic policy review and multistakeholder assessment in the future could complement our results.

Substrate prediction for RiPP biosynthetic enzymes via masked language modeling and transfer learning.

Ribosomally synthesized and post-translationally modified peptide (RiPP) biosynthetic enzymes often exhibit promiscuous substrate preferences that cannot be reduced to simple rules. Large language models are promising tools for predicting the specificity of RiPP biosynthetic enzymes. However, state-of-the-art protein language models are trained on relatively few peptide sequences. A previous study comprehensively profiled the peptide substrate preferences of LazBF (a two-component serine dehydratase) and LazDEF (a three-component azole synthetase) from the lactazole biosynthetic pathway. We demonstrated that masked language modeling of LazBF substrate preferences produced language model embeddings that improved downstream prediction of both LazBF and LazDEF substrates. Similarly, masked language modeling of LazDEF substrate preferences produced embeddings that improved prediction of both LazBF and LazDEF substrates. Our results suggest that the models learned functional forms that are transferable between distinct enzymatic transformations that act within the same biosynthetic pathway. We found that a single high-quality data set of substrates and non-substrates for a RiPP biosynthetic enzyme improved substrate prediction for distinct enzymes in data-scarce scenarios. We then fine-tuned models on each data set and showed that the fine-tuned models provided interpretable insight that we anticipate will facilitate the design of substrate libraries that are compatible with desired RiPP biosynthetic pathways.

Predicting College Closures and Financial Distress

In this paper, we assemble the most comprehensive dataset to date on the characteristics of colleges and universities, including dates of operation, institutional setting, student body, staff, and finance data from 2002 to 2023. We provide an extensive description of what is known and unknown about closed colleges compared with institutions that did not close. Using this data, we first develop a series of predictive models of financial distress, utilizing factors like operational revenue/expense patterns, sources of revenue, metrics of liquidity and leverage, enrollment/staff patterns, and prior signs of significant financial strain. We benchmark these models against existing federal government screening mechanisms such as financial responsibility scores and heightened cash monitoring. We document a high degree of missing data among colleges that eventually close and show that this is a key impediment to identifying at risk institutions. We then show that modern machine learning techniques, combined with richer data, are far more effective at predicting college closures than linear probability models, and considerably more effective than existing accountability metrics. Our preferred model, which combines an off-the-shelf machine learning algorithm with the richest set of explanatory variables, can significantly improve predictive accuracy even for institutions with complete data, but is particularly helpful for predicting instances of financial distress for institutions with spotty data. Finally, we conduct simulations using our estimates to contemplate likely increases in future closures, showing that enrollment challenges resulting from an impending demographic cliff are likely to significantly increase annual college closures for reasonable scenarios.

Evaluating AI and Machine Learning Models in Breast Cancer Detection: A Review of Convolutional Neural Networks (CNN) and Global Research Trends

Numerous studies have highlighted the significance of artificial intelligence (AI) in breast cancer diagnosis. However, systematic reviews of AI applications in this field often lack cohesion, with each study adopting a unique approach. The aim of this study is to provide a detailed examination of AI's role in breast cancer diagnosis through citation analysis, helping to categorize the key areas that attract academic attention. It also includes a thematic analysis to identify the specific research topics within each category. A total of 30,200 studies related to breast cancer and AI, published between 2015 and 2024, were sourced from databases such as IEEE, Scopus, PubMed, Springer, and Google Scholar. After applying inclusion and exclusion criteria, 32 relevant studies were identified. Most of these studies utilized classification models for breast cancer prediction, with high accuracy being the most commonly reported performance metric. Convolutional Neural Networks (CNN) emerged as the preferred model in many studies. The findings indicate that both the quantity and quality of AI-based algorithms in breast cancer diagnosis are increases in the given years. AI is increasingly seen as a complement to healthcare sector and clinical expertise, with the target of enhancing the accessibility and affordability of quality healthcare worldwide.

Rat Models of Breast Cancer.

As the first mammal to be domesticated for research purposes, rats served as the primary animal model for various branches of biomedical research, including breast cancer studies, up until the late 1990s and early 2000s. During this time, genetic engineering of mice, but not rats, became routine, and mice gradually supplanted rats as the preferred rodent model. But recent advances in creating genetically engineered rat models, especially with the assistance of CRISPR/Cas9 technology, have rekindled the significance of rats as a critical model in exploring various facets of breast cancer research. This is particularly pronounced in the study of the formation and progression of the estrogen receptor-positive subtype, which remains challenging to model in mice. In this chapter, we embark on a historical journey through the evolution of rat models in biomedical research and provide an overview of the general and histological characteristics of rat mammary glands. Next, we critically review major rat models for breast cancer research, including those induced by carcinogens, hormones, radiation, germline transgenes, germline knockouts, and intraductal injection of retrovirus/lentivirus to deliver oncogenic drivers into mature mammary glands. We also discuss the advances in building rat models using somatic genome editing powered by CRISPR/Cas9. This chapter concludes with our forward-looking perspective on future applications of advanced rat models in critical areas of breast cancer research that have continued to challenge the mouse model community.

Animal models in biomedical research: Relevance of Drosophila melanogaster.

Animal models have become veritable tools in gaining insight into the pathogenesis and progression of several human diseases. These models could range in complexity from Caenorhabditis elegans to non-human primates. With the aid of these animal models, a lot of new knowledge has been gained about several diseases which otherwise would not have been possible. Most times, the utilization of these animal models is predicated on the level of homology they share with humans, which suggests that outcomes of studies using them could be extrapolated to humans. However, this has not always been the case. Drosophila melanogaster is becoming increasingly relevant as preferred model for understanding the biochemical basis of several human diseases. Apart from its relatively short lifespan, high fecundity and ease of rearing, the simplicity of its genome and lower redundancy of its genes when compared with vertebrate models, as well as availability of genetic tool kit for easy manipulation of its genome, have all contributed to its emergence as a valid animal model of human diseases. This review aimed at highlighting the contributions of selected animal models in biomedical research with a focus on the relevance of Drosophila melanogaster in understanding the biochemical basis of some diseases that have continued to plague mankind.

백두대간 산림경관의 시각적 선호도 분석 연구

The purpose of this study is to find alternatives for improving the visual resource of forest resources and the landscape value in terms of public interest of forests through the analysis of visual preference of forest landscapes targeting Baekdudaegan forest resources. The research methods consist of theoretical consideration and presentation of an analysis framework, a visual preference survey for improving forest landscapes, and suggestions for improvement alternatives through this. The visual preference survey selected a sample group consisting of 50 landscape experts and used the SD scale technique using landscape adjectives. Then, the final preference model was presented through factor analysis and regression analysis. The implications of this study are that the selection of viewpoints and analysis of visual preferences in forest landscapes are meaningful results in terms of providing high ecosystem services through visual resource utilization of forest landscapes in terms of wise use of forest landscapes. In addition, it is judged to be significant as a basic research and reference frame when conducting in-depth research in terms of increasing the usability and efficient management of forest landscapes in the future.

Revisiting the Rewarding Properties and Abuse Potential of Natural Compound Sinomenine Using Conditioned Place Preference (CPP) Model in Mice

Sinomenine, derived from Sinomenium acutum, has a chemical structure similar to morphine and shows potential for treating opioid addiction in mice. Despite claims of being non- addictive, its safety and abuse potential remain underexplored. Conditioned place preference (CPP) approach was used to evaluate sinomenine's rewarding properties. ICR mice were divided into five groups, receiving either saline, morphine (8 mg/kg), or sinomenine (50, 100, or 200 mg/kg) where each group was confined to the drug-paired compartments during conditioning phase. Alternately, mice received saline when confined in the saline-paired compartment. During pre-conditioning and post-conditioning phases, the time spent in the drug-paired compartment, the number of entries to the drug-paired compartment, and rearing behaviours to assess the rewarding properties. Mice conditioned with morphine showed a significant increase in time spent in the drug-paired compartment from pre-conditioning to post-conditioning, but no significant changes in the number of entries and rearing behaviours. For saline and all sinomenine groups, there were no significant differences in the time spent, number of entries, and rearing behaviours except for sinomenine 200 mg/kg group, which showed a significant increase in rearing behaviours indicating an increase in exploratory behaviours. This study suggests that sinomenine, at various doses, is less likely to produce rewarding effects, indicating low abuse potential. The observed increase in exploratory behaviours during post-conditioning with sinomenine 200 mg/kg warrants further investigation.

Work-Life Balance and Work-Life Integration: A Comparative Analysis through Conceptual Distinction

The evolving discourse surrounding work-life balance and work-life integration reflects significant shifts in workplace dynamics, driven by factors such as increased workforce diversity, particularly the participation of women, and advancements in technology. This paper explores the conceptual distinctions between work-life balance, which emphasizes clear boundaries between professional and personal life, and work-life integration, which promotes the blending of work and non-work domains to create synergy. Work-life balance traditionally involves setting firm limits between paid work and personal responsibilities to maintain equilibrium, while work-life integration encourages flexibility, allowing individuals to fluidly navigate between professional and personal roles. Drawing on recent research, this paper highlights how each approach caters to different organizational structures and employee needs. Work-life integration, supported by the rise of remote work and flexible schedules, is especially suited for knowledge workers who benefit from autonomy. Conversely, work-life balance remains a preferred model for industries with fixed schedules or physical presence requirements, particularly among blue-collar workers. This article concludes by addressing the implications of both approaches for employee well-being, productivity, and organizational performance, emphasizing the need for tailored solutions to meet diverse workforce demands in today’s rapidly changing work environments.

Financial analysts’ use of industry specific stock valuation models

PurposeThe purpose of this paper is to explore the industry-specific preferred valuation model utilised by analysts’ in determining a stock’s target price. By understanding analysts’ use of industry-specific valuation models, we can enhance our comprehension of important aspects of value creation in these sectors. Therefore, understanding the industry context is crucial for accurately assessing the value of companies within that industry and selecting the most suitable valuation model.Design/methodology/approachThe method employed in this study is content analysis, examining the output of analysts’ valuation models within 25 Global Industry Classification Standard (GICS) industry groups. I hand-collected 806 equity reports from Capital IQ, selecting the four companies with the largest market capitalization from each of the 25 industry groups.FindingsPrice/Earnings (P/E) emerges as the preferred valuation model in 20 out of the 25 industry groups based on the GICS, with some exceptions. Notably, EV/EBITDA is favoured in the telecom, energy and materials sectors, while the capital goods industry primarily relies on Price/Cash flow (P/CF). In the Real Estate Investment Trusts (REITs) sector, P/AFFO (adjusted funds from operations) is the most commonly employed model. While earnings multiples remain the favoured valuation model for financial analysts, a noticeable shift away from multiperiod valuation models is evident after the first decade of the 21st century.Research limitations/implicationsThe findings can increase our comprehension of the interplay between valuation methodologies, industry characteristics and investment decision-making.Practical implicationsIt establishes a foundation for future research in this field and is anticipated to be of interest to analysts, fund managers and investors. The findings can increase our comprehension of the interplay between valuation methodologies, industry characteristics and investment decision-making.Originality/valueThis paper represents the first systematic and comprehensive examination of analysts’ utilisation of industry-specific stock valuation methods across all 25 GICS industry groups.

Neutral fraction of hydrogen in the intergalactic medium surrounding high-redshift gamma-ray burst 210905A

ABSTRACT The Epoch of Reionization (EoR) is a key period of cosmological history in which the intergalactic medium (IGM) underwent a major phase change from being neutral to almost completely ionized. Gamma-ray bursts (GRBs) are luminous and unique probes of their environments that can be used to study the timeline for the progression of the EoR. Here, we present a detailed analysis of the European Southern Observatory Very Large Telescope X-shooter spectrum of GRB 210905A, which resides at a redshift of $z\sim 6.3$. We focus on estimating the fraction of neutral hydrogen, $x_{\rm H \, {\small I}}$, on the line of sight to the host galaxy of GRB 210905A by fitting the shape of the Lyman-$\alpha$ damping wing of the afterglow spectrum. The X-shooter spectrum has a high signal-to-noise ratio, but the complex velocity structure of the host galaxy limits the precision of our conclusions. The statistically preferred model suggests a low neutral fraction with a 3$\sigma$ upper limit of $x_{\rm H \, {\small I}} \lesssim 0.15$ or $x_{\rm H \, {\small I}} \lesssim 0.23$, depending on the absence or presence of an ionized bubble around the GRB host galaxy, indicating that the IGM around the GRB host galaxy is mostly ionized. We discuss complications in current analyses and potential avenues for future studies of the progression of the EoR and its evolution with redshift.

A Framework for High-Spatiotemporal-Resolution Soil Moisture Retrieval in China Using Multi-Source Remote Sensing Data

High-spatiotemporal-resolution and accurate soil moisture (SM) data are crucial for investigating climate, hydrology, and agriculture. Existing SM products do not yet meet the demands for high spatiotemporal resolution. The objective is to develop and evaluate a retrieval framework to derive SM estimates with high spatial (100 m) and temporal (&lt;3 days) resolution that can be used on a national scale in China. Therefore, this study integrates multi-source data, including optical remote sensing (RS) data from Sentinel-2 and Landsat-7/8/9, synthetic aperture radar (SAR) data from Sentinel-1, and auxiliary data. Four machine learning and deep learning algorithms are applied, including Random Forest Regression (RFR), Extreme Gradient Boosting (XGBoost), Long Short-Term Memory (LSTM) networks, and Ensemble Learning (EL). The integrated framework (IF) considers three feature scenarios (SC1: optical RS + auxiliary data, SC2: SAR + auxiliary data, SC3: optical RS + SAR + auxiliary data), encompassing a total of 33 features. The results are as follows: (1) The correlation coefficients (r) between auxiliary data (such as sand fraction, r = −0.48; silt fraction, r = 0.47; and evapotranspiration, r = −0.42), SAR features (such as the backscatter coefficients for VV-pol (σvv0), r = 0.47), and optical RS features (such as Shortwave Infrared Band 2 (SWIR2) reflectance data from Sentinel-2 and Landsat-7/8/9, r = −0.39) with observed SM are significant. This indicates that multi-source data can provide complementary information for SM monitoring. (2) Compared to XGBoost and LSTM, RFR and EL demonstrate superior overall performance and are the preferred models for SM prediction. Their R2 for the training and test sets exceed 0.969 and 0.743, respectively, and their ubRMSE are below 0.022 and 0.063 m3/m3, respectively. (3) The SM prediction accuracy is highest for the scenario of optical + SAR + auxiliary data, followed by SAR + auxiliary data, and finally optical + auxiliary data. (4) With an increasing Normalized Difference Vegetation Index (NDVI) and SM values, the trained models exhibit a general decrease in prediction performance and accuracy. (5) In 2021 and 2022, without considering cloud cover, the IF theoretically achieved an SM revisit time of 1–3 days across 95.01% and 96.53% of China’s area, respectively. However, SC1 was able to achieve a revisit time of 1–3 days over 60.73% of China’s area in 2021 and 69.36% in 2022, while the area covered by SC2 and SC3 at this revisit time accounted for less than 1% of China’s total area. This study validates the effectiveness of combining multi-source RS data with auxiliary data in large-scale SM monitoring and provides new methods for improving SM retrieval accuracy and spatiotemporal coverage.

Self-Rated Symptoms of Oppositional Defiant Disorder and Conduct Disorder: Factor Structure, Reliability, and Validity in a Clinical Sample of Adolescents.

This study examined the psychometric properties of self-rated symptoms of oppositional defiant disorder (ODD) and conduct disorder (CD). We analyzed symptom ratings from a clinical outpatient sample of 658 adolescents aged 11;0-17;11 years diagnosed with ODD/CD and/or attention-deficit hyperactivity disorder. In two steps, we examined the factor structure of ODD models and an additional CD symptom dimension using confirmatory factor analysis (CFA) and exploratory structural equation modeling (ESEM). Our preferred three-factor CFA and ESEM models (CFI = 0.937-0.975, TLI = 0.926-0.962, RMSEA = 0.049-0.068, SRMR = 0.045-0.070) demonstrated scalar measurement invariance across age groups (≥ 14;0 years) and metric measurement invariance across informants (parents, teachers). With exceptions, the self-rated symptom scales showed acceptable internal consistencies (α = 0.68-0.84). Convergent (r =.27-.65) and divergent validity (r =.13-.41) were supported using Achenbach's Youth Self Report Form. Our results provide implications for using self-rated ODD/CD symptoms in clinical practice and research.

Edge Computing Enabled Lightweight Neural Network for User Preference Based Customization Recommendation Mechanism

ABSTRACTThe rapid development of computer and network technologies has led to an explosion of information, ushering in an era of e‐commerce dominated by online shopping. This shift underscores the need for precise, efficient, and customized recommendation systems. Traditional search engines struggle to cater to users' diverse information needs, leading to information overload and “dark information.” The customization recommendation systems emerge as a solution, leveraging user data to discern product correlations and formulate recommendations. However, remote server latency poses a challenge for real‐time recommendations. In this regard, we necessitate the exploration of edge computing and propose an edge computing enabled lightweight neural network for a user preference‐based customization recommendation mechanism including the edge computing based user preference model, the customization preference relevance, and the lightweight neural network‐based customization recommendation. In particular, the user preference model quantifies user affinity toward products and a preference connection pattern that ensures a clean relationship map by mitigating attribute interference, while the customization recommendation integrates back‐propagation for nonlinear input–output mappings, achieving generalization while maintaining efficiency. Experimental results indicate that the proposed mechanism can enhance customization and efficiency in recommendation systems when compared to the state‐of‐the‐art methods in terms of the recommendation accuracy, error rate as well as iterations.