Search

This study employs LendingClub data in the field of personal credit risk control as an illustrative case. Various data mining models, and support vector machine, are utilized for training purposes. Additionally, a Stacking model is integrated into the analysis to forecast customer default likelihood. Subsequently, lending decisions are made in accordance with these predictions. The outcomes indicate a reduction in customer default rates compared to scenarios without the application of data mining models, thereby achieving our goal of risk control. [ABSTRACT FROM AUTHOR]

An intelligent prediction model for rice yield in small-scale cultivation areas can provide precise forecasting results for farmers, rice planting enterprises, and researchers, holding significant importance for agricultural industries and crop science research within small regions. Although machine learning can handle complex nonlinear problems to enhance prediction accuracy, further improvements in models are still needed to accurately predict rice yields in small areas facing complex planting environments, thereby enhancing model performance. This study employs four rice phenotypic traits, namely, panicle angle, panicle length, total branch length, and grain number, along with seven machine learning methods—multiple linear regression, support vector machine, MLP, random forest, GBR, XGBoost, and LightGBM—to construct a yield prediction model group. Subsequently, the top three models with the best performance in individual model predictions are integrated using voting and stacking ensemble methods to obtain the optimal integrated model. Finally, the impact of different rice phenotypic traits on the performance of the stacked ensemble model is explored. Experimental results indicate that the random forest model performs best after individual machine learning modeling, with RMSE, R2, and MAPE values of 0.2777, 0.9062, and 17.04%, respectively. After model integration, Stacking–3m demonstrates the best performance, with RMSE, R2, and MAPE values of 0.2483, 0.9250, and 6.90%, respectively. Compared to the performance after random forest modeling, the RMSE decreased by 10.58%, R2 increased by 1.88%, and MAPE decreased by 0.76%, indicating improved model performance after stacking ensemble. The Stacking–3m model, which demonstrated the best comprehensive evaluation metrics, was selected for model validation, and the validation results were satisfactory, with MAE, R2, and MAPE values of 8.3384, 0.9285, and 0.2689, respectively. The above research findings demonstrate that this integrated model possesses high practical value and fills a gap in precise yield prediction for small-scale rice cultivation in the Yunnan Plateau region. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Computer Engineering & Applications is the property of Beijing Journal of Computer Engineering & Applications Journal Co Ltd. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

The International Maritime Organization (IMO) sets ambitious greenhouse gas reduction targets for the maritime industry. From a long-term net zero emission perspective, ammonia fuel is expected to play a significant role in the marine decarbonization journey compared to LNG as a transition fuel. Also, in addition to internal combustion engine applications, solid oxide fuel cells (SOFCs) have gained more attention in marine propulsion applications due to their high efficiency. This study was performed to investigate the technical feasibility of utilizing a closed-loop SOFC thermal energy release for ammonia decomposition, leading to hydrogen fuel generation and subsequently feed back into SOFCs. The result proves that the integrated system of ammonia cracking SOFCs can maintain a self-balanced condition, ensuring adequate SOFC heat supply for the ammonia cracking process to produce hydrogen while supporting normal SOFC operation and generating heat. This paper concludes that an integrated system represents a novel and feasible solution and emphasizes its potential as an adaptable solution for future marine applications.

Abstract Identifying the optimal mining methods plays a pivotal role in ensuring both economic efficiency and environmental sustainability. This study aims to propose a model that combines interval-valued Pythagorean fuzzy sets (IVPFS) and TOPSIS-GRA to select the optimal mining method for broken ore bodies. First, a multi-factor comprehensive evaluation system, including economic, safety, and technical aspects, was established. IVPFS was introduced to express the fuzzy information of the decision-making process within the evaluation system. Additionally, an objective method combining the principle of fuzzy entropy measurement with EWM was proposed to determine the weights of fuzzy information. This method distinguished the importance of decision-makers and indicators. Then, an integration of distance and similarity (TOPSIS-GRA) was employed for ranking alternative solutions to select the optimal one. This model was applied to the decision-making problem of mining methods for the broken and difficult-to-mine ore bodies in the Tanyaokou mining area. Initial fuzzy evaluation information was obtained by having decision-makers score the mining methods. Results showed that the comprehensive scores of four alternatives are 0.5172, 0.4683, 0.5192, and 0.5465, respectively. The optimal method was the point-pillar upward horizontal layered filling mining method. Finally, the sensitivity analysis confirmed the stability of the model. The comparative results under different fuzzy environments (PFS and TFS) demonstrated the strong capability of IVPFS in handling fuzzy information for optimizing mining methods.

Ladle furnace (LF) refining can effectively control the composition and temperature of molten steel and plays a role in cushioning and coordinating the production rhythm between steelmaking and continuous casting. The use of models for control and decision-making in LF refining can further standardize the refining operations, improve the quality and stability of molten steel, and, combined with automatic control, will strongly promote the development of intelligent refining to achieve optimization of steelmaking and improve efficiency. Regarding promoting intelligent manufacturing in the steel industry, the LF refining process model is no longer limited to the establishment and deployment of single-function models and has begun to develop in the direction of integration, automation, and intelligence while its function has also changed from a single prediction and recommendation to overall intelligent control and decision-making. LF process control and decision models are mostly single-function models, but few integrate applications. Due to the complexity and uncertainty of the refining process, these models have differences in stability and accuracy. Therefore, establishing an integrated model, standardizing the field process, improving the data quality, and combining automatic control and closed-loop feedback to further realize the intelligent control model have become important directions for future research and application of LF control models. Herein, the development and research status of LF refining control and decision models are summarized, including the alloying model, slagging model, temperature model, argon blowing control model, calcium treatment model, and other single-function models, as well as intelligent refining technology. The modeling principles and functions of these models are systematically reviewed, and future development directions of LF process intelligent control and decision models are prospected, providing a reference for the subsequent development and application of LF intelligent refining technology. The establishment and real landing of LF intelligent control and decision models not only require the realization and linkage of process control and decision models but also propose higher requirements for iron and steel enterprises. The realization of LF intelligent control and decision-making models can greatly improve the consistency and qualified rate of product quality, reduce energy consumption and cost, reduce manual intervention, and shorten the smelting cycle, thus improving the competitiveness of enterprises. With the continuous upgrading and improvement of model design, automation technology, and steel mill site environment, the application and development of LF intelligent control and decision-making models show great potential in realizing green, low-carbon, and intelligent manufacturing and would make great contributions to the progress and transformation and upgrading of the steel industry in the future.

Abstract The detection of landslide areas and surface characteristics is the prerequisite and basis of landslide hazard risk assessment. The traditional method relies mainly on manual field identification, and discrimination is based on the lack of unified quantitative standards. Thus, the use of neural networks for the quantitative identification and prediction of landslide surface deformation is explored. By constructing an integrated model based on YOLO X-CNN and Mask R-CNN, a deep learning-based feature detection method for landslide surface images is proposed. First, the method superimposes Unmanned Aerial Vehicle (UAV) oblique photography data (UOPD) and Internet heterosource image data (IHID) to construct a landslide surface image dataset and landslide surface deformation database. Second, an integrated model suitable for small- and medium-scale target detection and large-scale target edge extraction is constructed to automatically identify and extract landslide surface features and to achieve rapid detection of landslide surface features and accurate segmentation and deformation recognition of landslide areas. The results show that the detection accuracy for small rock targets is greater than 80% and that the speed is 57.04 FPS. The classification and mask segmentation accuracies of large slope targets are approximately 90%. A speed of 7.89 FPS can meet the needs of disaster emergency response; this provides a reference method for the accurate identification of landslide surface features.

Abstract Background Accurate differentiation between malignant and benign pulmonary nodules, especially those measuring 5–10 mm in diameter, continues to pose a significant diagnostic challenge. This study introduces a novel, precise approach by integrating circulating cell-free DNA (cfDNA) methylation patterns, protein profiling, and computed tomography (CT) imaging features to enhance the classification of pulmonary nodules. Methods Blood samples were collected from 419 participants diagnosed with pulmonary nodules ranging from 5 to 30 mm in size, before any disease-altering procedures such as treatment or surgical intervention. High-throughput bisulfite sequencing was used to conduct DNA methylation profiling, while protein profiling was performed utilizing the Olink proximity extension assay. The dataset was divided into a training set and an independent test set. The training set included 162 matched cases of benign and malignant nodules, balanced for sex and age. In contrast, the test set consisted of 46 benign and 49 malignant nodules. By effectively integrating both molecular (DNA methylation and protein profiling) and CT imaging parameters, a sophisticated deep learning-based classifier was developed to accurately distinguish between benign and malignant pulmonary nodules. Results Our results demonstrate that the integrated model is both accurate and robust in distinguishing between benign and malignant pulmonary nodules. It achieved an AUC score 0.925 (sensitivity = 83.7%, specificity = 82.6%) in classifying test set. The performance of the integrated model was significantly higher than that of individual methylation (AUC = 0.799, P = 0.004), protein (AUC = 0.846, P = 0.009), and imaging models (AUC = 0.866, P = 0.01). Importantly, the integrated model achieved a higher AUC of 0.951 (sensitivity = 83.9%, specificity = 89.7%) in 5–10 mm small nodules. These results collectively confirm the accuracy and robustness of our model in detecting malignant nodules from benign ones. Conclusions Our study presents a promising noninvasive approach to distinguish the malignancy of pulmonary nodules using multiple molecular and imaging features, which has the potential to assist in clinical decision-making. Trial registration: This study was registered on ClinicalTrials.gov on 01/01/2020 (NCT05432128). https://classic.clinicaltrials.gov/ct2/show/NCT05432128 .

A modern approach to personalized recommendation systems is presented, combining graph neural networks GNN with RL reinforcement learning methods. The GNN model is optimized for recommendation systems and is trained on vector representations of users and products, which are used to generate an initial list of recommendations that are fed into the RL model. Particular attention is paid to the architecture and operation of the integrated GNN-RL model. The results of experimental studies demonstrating the effectiveness of the proposed approach are presented. [ABSTRACT FROM AUTHOR]

Management boundaries are often delimited by political and social factors, whereas animal movements are affected by ecological and geophysical constraints. Thus, understanding connectivity among distinct management units is of considerable importance, particularly for harvested species, where quotas set in ignorance of connectivity may fail to meet management goals.We constructed an individual‐based model (IBM) to better understand wild turkey movements at large scales, benefiting from multiple data sources that are often available for harvested species.We built an IBM from spring seasonal movements of wild turkey, using data from ringed, radio‐, and GPS‐marked turkeys captured in Maine, USA. Our IBM accommodated variation in individual turkey response to landscape connectivity metrics and identified emergent migratory connectivity dynamics among harvest management regions.We calculated a low degree of connectivity among wildlife management districts (WMD) which, in combination with the substantial number of boundary crossings observed, indicated a more diffuse distribution of turkeys among WMDs.Synthesis and applications: Estimates of turkeys moving between districts provided a clear delineation of where immigration was strongest, identifying which WMDs should be managed as singular population units. This approach has widespread utility for any species or system where harvest management decisions are made at finer spatial scales than the movement dynamics affecting population processes. [ABSTRACT FROM AUTHOR]

The International Maritime Organization (IMO) sets ambitious greenhouse gas reduction targets for the maritime industry. From a long-term net zero emission perspective, ammonia fuel is expected to play a significant role in the marine decarbonization journey compared to LNG as a transition fuel. Also, in addition to internal combustion engine applications, solid oxide fuel cells (SOFCs) have gained more attention in marine propulsion applications due to their high efficiency. This study was performed to investigate the technical feasibility of utilizing a closed-loop SOFC thermal energy release for ammonia decomposition, leading to hydrogen fuel generation and subsequently feed back into SOFCs. The result proves that the integrated system of ammonia cracking SOFCs can maintain a self-balanced condition, ensuring adequate SOFC heat supply for the ammonia cracking process to produce hydrogen while supporting normal SOFC operation and generating heat. This paper concludes that an integrated system represents a novel and feasible solution and emphasizes its potential as an adaptable solution for future marine applications. [ABSTRACT FROM AUTHOR]

Hepatocellular carcinoma (HCC) is a deadly cancer that emerges from a continuous progression of liver cells from normal to abnormal, often following infections by hepatitis B/C viruses (HBV/HCV), liver fibrosis, and liver cirrhosis (LC), ultimately culminating in cancer. However, there is currently limited systematic molecular analysis of biomarkers at different stages of HCC progression using multi‐omics approaches. We carried out an innovative pipeline by utilizing targeted proteomics and metabolomics to identify potential biomarkers for early detection of HCC in 316 participants, including healthy adults and patients diagnosed with HBV, HCV, LC, and HCC from three independent cohorts. We first established a detailed database of candidate biomarkers for HCC containing 3059 proteins and 103 metabolites, and identified pivotal candidates implicated in the progressive trajectory of liver cancers. Through our developed DeepPRM, scheduled multiple reaction monitoring (MRM)‐targeted approach, and machine learning‐based computational pipeline, we identified an eight‐biomolecular‐based combination with an accuracy rate of 91.43% for early diagnosis of HCC, and a 12‐biomolecular‐based combination with an accuracy rate of 80.00% for detecting changes in HBV–LC progression. These two biomarker combinations significantly improved accuracy compared to traditional tumor biomarkers. Our extensive analysis provides valuable proteomic and metabolomic data resources that will contribute to a deeper understanding of liver disease progression and enhance the identification of potential therapeutic targets. [ABSTRACT FROM AUTHOR]

The efficiency of the management of predations on livestock by gray wolves (Canis lupus) through culling is under debate. Evaluating wolf culling efficiency requires to simultaneously analyze the effects of culling on the wolf population and the repercussions of these population changes on livestock predation. This protocol is technically difficult to implement in the field. To properly assess culling efficiency, we provided an integrated and flexible individual‐based model that simulated interactions between wolf population dynamics, predation behavior and culling management. We considered many social processes in wolves. We calibrated the model to match the Western Alps as a case study. By considering the prey community in this area and the opportunistic nature of wolf predation, we assumed that predation on livestock at the wolf territory level increased with pack's food needs. Under this assumption and considering livestock availability as high and livestock vulnerability as uniform in space and time, culling maintained wolf population size and predation risks at low levels. Contrary to what was expected, culling decreased the mean annual proportions of dispersing wolves in our simulations, by speeding settlement. This population‐level mechanism compensated for the high mortality and the pack instability caused by culling. Compensation was however dependent on the selectivity and the timing of culling. When executed before the natural mortality module in our model, the selective culling could undermine replacement of lost breeders and therefore decrease wolf population resilience to culling. Our model gives insights about culling effects in one specific simulated context, but we do not expect that our assumption about predation behavior necessarily holds in other ecological contexts and we therefore encourage further explorations of the model. [ABSTRACT FROM AUTHOR]

Purpose: This study aims to expand antecedent roles on social entrepreneurial behavioural intention by integrating both the Three Pillars of Institutions and the Mair Noboa model. The literature lacks in investigating both institutional- and individual-level antecedents to determine social entrepreneurial behavioural intention. This proposed integrated model was developed in which the Mair Noboa's model antecedents mediates the positive relationship between the antecedents of Three Pillars of Institutions and social entrepreneurial intention. Design/methodology/approach: This study uses quantitative research methodologies to answer the research question of the extent that institutional-level antecedents in turn influence individual antecedents and thus determine social entrepreneurial intention. To explore this, a Web-based survey distributed across Bangladesh (n = 412). The confirmation of hypotheses involved using covariance-based structural equation modelling (SEM) for data analysis. The resulting measurement and structural models successfully met all criteria for reliability, model fit, convergent validity and discriminant validity. The hypotheses were subsequently assessed by examining both direct relationships and mediating effects. Findings: The findings demonstrated a significant relationship between the antecedents of the Three Pillars of Institutions and the Mair Noboa model. The results suggest that the Mair Noboa model antecedents can mediate the relationship between the Three Pillars of Institutions and social entrepreneurial intention. Originality/value: This paper advances the existing knowledge of social entrepreneurial intention, through the novel lens of combined institutional and individual antecedents. This paper fills an important knowledge gap by exploring both institutional- and individual-level antecedents to determine social entrepreneurial intention. This study findings yield fresh theoretical and practical insights into how institutional and individual antecedents jointly influence social entrepreneurial intention. [ABSTRACT FROM AUTHOR]

Human errors (HEs) are prevalent issues in manual assembly, leading to product defects and increased costs. Understanding and knowing the factors influencing human errors in manual assembly processes is essential for improving product quality and efficiency. This study aims to determine and rank factors influencing HEs in manual assembly processes based on expert judgments. To achieve this objective, an integrated model was developed using two multi-criteria decision-making (MCDM) techniques—specifically, the fuzzy Delphi Method (FDM) and the fuzzy Analytic Hierarchy Process (FAHP). Firstly, two rounds of the FDM were conducted to identify and categorize the primary factors contributing to HEs in manual assembly. Expert consensus with at least 75% agreement determined that 27 factors with influence scores of 0.7 or higher significantly impact HEs in these processes. After that, the priorities of the 27 influencing factors in assembly HEs were determined using a third round of the FAHP method. Data analysis was performed using SPSS 22.0 to evaluate the reliability and normality of the survey responses. This study has divided the affecting factors on assembly HEs into two levels: level 1, called main factors, and level 2, called sub-factors. Based on the final measured weights for level 1, the proposed model estimation results revealed that the most influential factors on HEs in a manual assembly are the individual factor, followed by the tool factor and the task factor. For level 2, the model results showed a lack of experience, poor instructions and procedures, and misunderstanding as the most critical factors influencing manual assembly errors. Sensitivity analysis was performed to determine how changes in model inputs or parameters affect final decisions to ensure reliable and practical results. The findings of this study provide valuable insights to help organizations develop effective strategies for reducing worker errors in manual assembly. Identifying the key and root factors contributing to assembly errors, this research offers a solid foundation for enhancing the overall quality of final products. [ABSTRACT FROM AUTHOR]

Background: Diabetes mellitus (DM) is an ongoing and growing health problem worldwide, with a series of important complications such as diabetic foot that can significatively reduce the quality of life of affected patients. This study aims to explore the socio-cultural aspects of patients with diabetic foot, analyzing the following research question: "What are the socio-cultural aspects experienced by patients with diabetic foot?" Methods: A qualitative design using an ethnographic approach was applied to study the social and cultural aspects of Italian diabetic foot patients. Results: We included 20 key informants: 13 men and 7 women. Ages ranged from 54 to 71, with an average age of 61.2. The data analysis revealed five main themes: perceptions of diabetic foot, living with diabetic foot, impacts of culture and economic performance, barriers to health and diabetic foot, and home remedies and alternative medicine. Conclusions: This study provides a new perspective on the influence of cultural factors on the health of diabetic foot patients, showing various factors related to a lack of knowledge and training, fear, and acceptance of diabetic foot. This study also presents a new integrated model which will allow patients and practitioners to act on the various critical issues that emerged from our research. [ABSTRACT FROM AUTHOR]

Background: Accurate differentiation between malignant and benign pulmonary nodules, especially those measuring 5–10 mm in diameter, continues to pose a significant diagnostic challenge. This study introduces a novel, precise approach by integrating circulating cell-free DNA (cfDNA) methylation patterns, protein profiling, and computed tomography (CT) imaging features to enhance the classification of pulmonary nodules. Methods: Blood samples were collected from 419 participants diagnosed with pulmonary nodules ranging from 5 to 30 mm in size, before any disease-altering procedures such as treatment or surgical intervention. High-throughput bisulfite sequencing was used to conduct DNA methylation profiling, while protein profiling was performed utilizing the Olink proximity extension assay. The dataset was divided into a training set and an independent test set. The training set included 162 matched cases of benign and malignant nodules, balanced for sex and age. In contrast, the test set consisted of 46 benign and 49 malignant nodules. By effectively integrating both molecular (DNA methylation and protein profiling) and CT imaging parameters, a sophisticated deep learning-based classifier was developed to accurately distinguish between benign and malignant pulmonary nodules. Results: Our results demonstrate that the integrated model is both accurate and robust in distinguishing between benign and malignant pulmonary nodules. It achieved an AUC score 0.925 (sensitivity = 83.7%, specificity = 82.6%) in classifying test set. The performance of the integrated model was significantly higher than that of individual methylation (AUC = 0.799, P = 0.004), protein (AUC = 0.846, P = 0.009), and imaging models (AUC = 0.866, P = 0.01). Importantly, the integrated model achieved a higher AUC of 0.951 (sensitivity = 83.9%, specificity = 89.7%) in 5–10 mm small nodules. These results collectively confirm the accuracy and robustness of our model in detecting malignant nodules from benign ones. Conclusions: Our study presents a promising noninvasive approach to distinguish the malignancy of pulmonary nodules using multiple molecular and imaging features, which has the potential to assist in clinical decision-making. Trial registration: This study was registered on ClinicalTrials.gov on 01/01/2020 (NCT05432128). https://classic.clinicaltrials.gov/ct2/show/NCT05432128. [ABSTRACT FROM AUTHOR]

Understanding the factors that drive consumers to purchase electric vehicles (EVs) is critical to achieving decarbonization of China's transportation sector, as well as mitigating global warming. This study aims to construct a research model based on altruistic and self-interested perspectives by integrating the Theory of Planned Behaviour (TPB) and Norm Activation Model (NAM) to predict the psychological factors that influence Chinese consumers' intention to purchase EVs. Data were collected from 867 participants in China and empirically tested using Structural Equation Modeling (SEM). Self-interested factors, namely subjective norms, attitudes and perceived behavioural control, all had a significant positive effect on EV purchase intention. Additionally, the results showed that personal norms had the greatest effect on EV purchase intention. It was also found that awareness of consequence, ascription of responsibility and subjective norms were positive predictors of personal norms. Awareness of consequence had a positive effect on both the ascription of responsibility and attitudes. The findings contribute to understanding the psychological drivers of Chinese consumers' intention to purchase EVs and can provide decision-making references for policy makers and manufacturers. [ABSTRACT FROM AUTHOR]

AbstractFunctional fermented foods are increasingly recognized for their positive impact on human health, and this study aims to identify the factors that influence these products’ consumption in Iran. The study uses structural equation modeling (SEM) to test the conceptual model incorporating the theory of planned behavior (TPB), protection motivation theory (PMT), and flow theory (FT). Data from 319 individuals having prior experience with such foods were analyzed to test the hypothesized relationships. Based on the results, the integrated model components demonstrated a strong predictive power, explaining 70.6% of the variance in functional fermented food consumption. The study found that perceived vulnerability, perceived severity, and response cost from the PMT were the main drivers of functional fermented food consumption among Iranians. The study also offers practical insights, including highlighting health benefits, promoting pleasurable aspects, and mitigating accessibility challenges to these foods for Iranian consumers. [ABSTRACT FROM AUTHOR]

We report on a quality improvement initiative to facilitate biopsychosocial approaches for medical patients in an acute hospital setting through a hybrid integrated psychology care model. The expectation was to improve patient outcomes by increasing provider satisfaction and reducing average length of stay (ALOS). Psychologists in the adult consultation–liaison (CL) service were embedded with two service lines: hematology–oncology and medical trauma teams to comanage medical patients in their daily care through an interdisciplinary integrated approach. After 6 months, we compared differences in the ALOS between the traditional CL and hybrid integrated models. Satisfaction with the psychology services among providers was evident with 97% noting that integrated psychologists would reduce their own burnout. ALOS for patients evaluated by psychologists in the CL service was not statistically significantly different from the hybrid integrated model (CL service ALOS = 28 days vs. hybrid integrated pilot model ALOS = 20 days, p =.603). Earlier psychology evaluations (i.e., conducted within 5 days of admission) resulted in statistically significantly lower LOS in both models (p ≤.002). An integrated approach to patient care showed the potential to reduce LOS especially when psychological evaluation occurred within 5 days of admission. Additionally, the integrated model resulted in improved staff satisfaction. This collaboration can be of significant clinical and potential monetary value for the medical field as a whole. Public Significance Statement: This study advances the notion that integrating a psychologist within a medical team in an acute care medical setting can improve overall hospital outcomes for both patients and physicians. Additionally, it highlights how to maximize efficiencies of health care services being utilized, which has significant clinical and potential monetary value for the medical field. Hence, this approach aligns with the quadruple aim of health care. [ABSTRACT FROM AUTHOR]

The development of Artificial Intelligence (AI) has significantly influenced how consumers search for information. However, there is a lack of comprehensive models based on theoretical foundations that specifically address AI-powered information search behavior. This study draws on psychological motivation, information processing, and information economics theories to develop a theoretical model of consumer AI-powered information search behavior. The study aims to identify the main factors affecting consumer search behavior, offering a more holistic understanding of consumer behavior in the context of AI. Analyzing 512 valid questionnaires, the study shows that search motivation not only had the most significant impact on search behavior but also served as a mediator between other variables and search intensity. Additionally, perceived search ability had a direct and the greatest indirect impact on search behavior, while other variables such as perceived search costs and benefits also had indirect effects on search behavior. Practically, the study offers valuable insights for businesses and AI developers. Understanding the factors that drive AI-powered search behavior can inform the design of more effective AI systems and marketing strategies. [ABSTRACT FROM AUTHOR]

The global economic turmoil and the employability gap among university students have created a challenging job market for graduates. It is crucial to understand the factors that shape students' perceived employability and to implement effective measures to address this serious issue. A literature review was conducted to provide a comprehensive understanding of the factors influencing perceived employability among university students, focusing on research findings published since 2021. The present study indicates that numerous factors, including individual, behavioural, and environmental variables, significantly influence perceived employability among university students. Besides, previous research has diverged on specific factors such as demographics, partial components of personality traits, career engagement, political skills, and labour market situation, which need to be further investigated in the future. Furthermore, this study underscores the crucial significance of perceived employability among university students in a challenging job market. It validates the relevance of Career Construction Theory in interpreting perceived employability among university students and improves researchers' understanding of how perceived employability has evolved before and after the pandemic. Most importantly, this study develops an integrated model of the factors influencing the perceived employability among university students, which can serve as a guide for future researchers to study university students' employment issues and navigate the career practices of stakeholders such as the Ministry of Education, universities, university students, employers, and career practitioners. [ABSTRACT FROM AUTHOR]

Abstract Background Multi‐omics features of cell‐free DNA (cfDNA) can effectively improve the performance of non‐invasive early diagnosis and prognosis of cancer. However, multimodal characterization of cfDNA remains technically challenging. Methods We developed a comprehensive multi‐omics solution (COMOS) to specifically obtain an extensive fragmentomics landscape, presented by breakpoint characteristics of nucleosomes, CpG islands, DNase clusters and enhancers, besides typical methylation, copy number alteration of cfDNA. The COMOS was tested on 214 plasma samples of diffuse large B‐cell lymphoma (DLBCL) and matched healthy controls. Results For early diagnosis, COMOS improved the area under the curve (AUC) value to .993 compared with the individual omics model, with a sensitivity of 95% at 98% specificity. Detection sensitivity achieved 91% at 99% specificity in early‐stage patients, while the AUC values of the individual omics model were 0.942, 0.968, 0.989, 0.935, 0.921, 0.781 and 0.917, respectively, with lower sensitivity and specificity. In the treatment response cohort, COMOS yielded a superior sensitivity of 88% at 86% specificity (AUC, 0.903). COMOS has achieved excellent performance in early diagnosis and treatment response prediction. Conclusions Our study provides an effectively improved approach with high accuracy for the diagnosis and prognosis of DLBCL, showing great potential for future clinical application. Key points A comprehensive multi‐omics solution to specifically obtain an extensive fragmentomics landscape, presented by breakpoint characteristics of nucleosomes, CpG islands, DNase clusters and enhancers, besides typical methylation, copy number alteration of cfDNA. Integrated model of cfDNA multi‐omics could be used for non‐invasive early diagnosis of DLBCL. Integrated model of cfDNA multi‐omics could effectively evaluate the efficacy of R‐CHOP before DLBCL treatment.

Relevance: the article analyzes an alternative view of the model of sustainable development of the pipe industry. A comparative description of the classical and author’s models is given. The following methods were used in the work: analysis of reliable sources, up-to-date data, as well as calculations of an integrated model of sustainable development using the example of the Chelyabinsk Pipe Rolling Plant. The scientific novelty of the research consists in the fact that the author’s model of sustainable development of the pipe industry is proposed, justifi ed, and substantiated on the basis of the interaction of the main areas of sustainability (on the example of the Russian Federation and the European Union). The results and conclusions of the article can be useful for the scientifi c community and pipe industry enterprises in the development of strategic economic development programs.

With the increasing frequency of extreme weather events and a deepening understanding of disasters, resilience has received widespread attention in urban drainage systems. The studies on the resilience assessment of urban drainage systems are mostly indirect assessments that did not simulate human behavior affected by rainfall or semi-quantitative assessments that did not build simulation models, but few research characterizes the processes between people and infrastructure to assess resilience directly. Our study developed a dynamic model that integrates urban mobility, flood inundation, and sewer hydrodynamics processes. The model can simulate the impact of rainfall on people's mobility behavior and the full process including runoff generation, runoff entering pipes, node overflow, flood migration, urban mobility, and residential water usage. Then, we assessed the resilience of the urban drainage system under rainfall events from the perspectives of property loss and urban mobility. The study found that the average percentage increase in commuting time under different return periods of rainfall ranged from 6.4 to 203.9%. Calculating the annual expectation of property loss and traffic obstruction, the study found that the annual expectation loss in urban mobility is 9.1% of the annual expectation of property loss if the rainfall is near the morning commuting peak. HIGHLIGHTS Assessed the resilience of the drainage system from the perspectives of property and urban.; Developed a model that integrates urban mobility, flood inundation, and sewer hydrodynamics processes.; The average percentage increase in commuting time ranged from 6.4 to 203.9%.; The annual expectation loss in urban mobility is 9.1% of the annual expectation of property loss if the rainfall is near the morning commuting peak.;

Accurate and timely risk assessment of short-term rainstorm-type flood disasters is very important for ecological environment protection and sustainable socio-economic development. Given the complexity and variability of different geographical environments and climate conditions, a single machine learning model may lead to overfitting issues in flood disaster assessment, limiting the generalization ability of such models. In order to overcome this challenge, this study proposed a short-term rainstorm flood disaster risk assessment framework under the integrated learning model, which is divided into two stages: The first stage uses microwave remote sensing images to extract flood coverage and establish disaster samples, and integrates multi-source heterogeneous data to build a flood disaster risk assessment index system. The second stage, under the constraints of Whale Optimization Algorithm (WOA), optimizes the integration of random forest (RF), support vector machine (SVM), and logistic regression (LR) base models, and then the WRSL-Short-Term Flood Risk Assessment Model is established. The experimental results show that the Area Under Curve (AUC) accuracy of the WRSL-Short-Term Flood Risk Assessment Model is 89.27%, which is 0.95%, 1.77%, 2.07%, 1.86%, and 0.47% higher than RF, SVM, LR, XGBoost, and average weight RF-SVM-LR, respectively. The accuracy evaluation metrics for accuracy, Recall, and F1 Score have improved by 5.84%, 21.50%, and 11.06%, respectively. In this paper, WRSL-Short-Term Flood Risk Assessment Model is used to carry out the risk assessment of flood and waterlogging disasters in Henan Province, and ArcGIS is used to complete the short-term rainstorm city flood and waterlogging risk map. The research results will provide a scientific assessment basis for short-term rainstorm city flood disaster risk assessment and provide technical support for regional flood control and risk management. [ABSTRACT FROM AUTHOR]

As the popularity of electric bicycles (e-bikes) continues to surge, the number of accidents involving them has commensurately increased. A significant factor contributing to the high fatality rate in these accidents is the low usage of helmets among e-bike riders. Helmets have been proven to reduce the severity of injuries, yet their usage remains unexpectedly low. This issue is particularly pronounced among college students, the primary buyer group for e-bikes. Regrettably, there is a lack of research exploring their intentions to wear helmets. Understanding determinants of their intentions to wear helmets is crucial in promoting safe e-bike travel. Therefore, the present study aims to develop an integrated theoretical model that combines the Theory of Planned Behavior (TPB) and the Health Belief Model (HBM) to examine the factors influencing e-bike riders' helmet-wearing intentions among college students. Additionally, two variables—descriptive norms and law enforcement—are incorporated. The results indicate that the integrated model accounts for 76% of the variance in helmet-wearing intention, surpassing single-theory models. Specifically, the TPB accounts for 65%, while the HBM explains 53%. Notably, law enforcement emerges as the most influential factor, highlighting the crucial role of enforcing regulations and promoting awareness. Other significant factors include subjective and descriptive norms, attitudes, perceived benefits, perceived susceptibility, perceived barriers, and perceived severity. These findings provide valuable insights for policy development and targeted interventions aimed at improving helmet wear rates among e-bike riders, especially among the college student population. [ABSTRACT FROM AUTHOR]

Due to the rapid population and economic growth, the demand for water has increased. In addition, the natural resources are limited and degrade because of several factors such as the climate change. These challenges lead to reduce the ability of providing water at the required quantity and quality. One of solutions to maintain the sustainability of water supply from different sources is reuse of wastewater. For this aim, it is crucial to optimize wastewater systems. This research paper aims to describe different modelling possibilities and optimization methods for various components of integrated urban wastewater systems. The main conclusion of this research paper is the lack of study of optimum design and operation of urban wastewater systems in a holistic method. Moreover, most of previous studies on integrated wastewater management have been conducted on combined sewer systems. [ABSTRACT FROM AUTHOR]

Liquefied natural gas (LNG), recognized as the fossil fuel with the lowest carbon emission intensity, is a crucial transitional energy source in the global shift towards low-carbon energy. As the natural gas industry undergoes rapid expansion, the complexity of investment business models has increased significantly. Optimizing the combination of various segments within the value chain has become standard practice, making it essential to control risks and enhance economic benefits in these multifaceted scenarios. This paper introduces an integrated economic model encompassing upstream, liquefaction, shipping, regasification, and consumption, suitable for both upstream and downstream integration. The model offers a comprehensive analysis of the primary business models and key factors across each segment of the value chain. By constructing a robust economic evaluation framework, the study aims to provide a holistic approach to understanding the economic feasibility of LNG projects. Two detailed case studies are conducted to demonstrate the application and effectiveness of the proposed model, highlighting its capability to guide investment decisions, support risk management, and optimize asset portfolios. The integrated economic model developed in this study serves as a valuable tool for stakeholders in the LNG industry. It not only facilitates informed investment decision-making but also enhances the strategic management of risks and resources. By leveraging this model, investors and managers can better navigate the complexities of the LNG business, ensuring sustainable and economically viable operations. [ABSTRACT FROM AUTHOR]

Quantifying temporal and spatial variation in animal population size and demography is a central theme in ecological research and important for directing management and policy. However, this requires field sampling at large spatial extents and over long periods of time, which is not only prohibitively costly but often politically untenable. Participatory monitoring programs (also called citizen science programmes) can alleviate these constraints by recruiting stakeholders and the public to increase the spatial and temporal resolution of sampling effort and hence resulting data. While the majority of participatory monitoring programs are limited by opportunistic sampling designs, we are starting to see the emergence of structured citizen science programs that employ trained volunteers to collect data according to standardized protocols. Simultaneously, there is much ongoing development of statistical models that are increasingly more powerful and able to make more efficient use of field data. Integrated population models (IPMs), for example, are able to use multiple streams of data from different field monitoring programmes and/or multiple aspects of single datasets to estimate population sizes and key vital rates. Here, we developed a multi-area version of a recently developed integrated distance sampling model (IDSM) and applied it to data from a large-scale participatory monitoring program – the “Hønsefuglportalen” – to study spatio-temporal variation in population dynamics of willow ptarmigan (Lagopus lagopus) in Norway. We constructed an open and reproducible workflow for exploring temporal, spatial (latitudinal, longitudinal, altitudinal), and residual variation in recruitment, survival, and population density, as well as relationships between vital rates and relevant covariates and signals of density dependence. Recruitment rates varied more across space than over time, while the opposite was the case for survival. Slower life history patterns (higher survival, lower recruitment) appeared to be more common at higher latitudes and altitudes, portending differential effects of climate change on ptarmigan across their range. While there was variation in the magnitude of the effect small rodent occupancy had on recruitment, the relationships were predominantly positive and thus consistent with the alternative prey hypothesis. Notably, the accurate estimation of covariate effect was only made possible by integrating data from several monitoring areas for analysis. Our study highlights the potential of participatory monitoring and 2integrated modelling approaches for estimating and understanding spatio-temporal patterns in species abundance and demographic rates, and showcases how corresponding workflows can be set up in reproducible and semi-automated ways that increase their usefulness for informing management and regular reporting towards national and international biodiversity frameworks.

Abstract Hepatocellular carcinoma (HCC) is a deadly cancer that emerges from a continuous progression of liver cells from normal to abnormal, often following infections by hepatitis B/C viruses (HBV/HCV), liver fibrosis, and liver cirrhosis (LC), ultimately culminating in cancer. However, there is currently limited systematic molecular analysis of biomarkers at different stages of HCC progression using multi‐omics approaches. We carried out an innovative pipeline by utilizing targeted proteomics and metabolomics to identify potential biomarkers for early detection of HCC in 316 participants, including healthy adults and patients diagnosed with HBV, HCV, LC, and HCC from three independent cohorts. We first established a detailed database of candidate biomarkers for HCC containing 3059 proteins and 103 metabolites, and identified pivotal candidates implicated in the progressive trajectory of liver cancers. Through our developed DeepPRM, scheduled multiple reaction monitoring (MRM)‐targeted approach, and machine learning‐based computational pipeline, we identified an eight‐biomolecular‐based combination with an accuracy rate of 91.43% for early diagnosis of HCC, and a 12‐biomolecular‐based combination with an accuracy rate of 80.00% for detecting changes in HBV–LC progression. These two biomarker combinations significantly improved accuracy compared to traditional tumor biomarkers. Our extensive analysis provides valuable proteomic and metabolomic data resources that will contribute to a deeper understanding of liver disease progression and enhance the identification of potential therapeutic targets.

Addressing the critical need for enhanced power usage effectiveness in data centers (DCs), this study pioneers an improved convolutional long short-term memory with deep neural network (CLDNN) model, enriched with attention mechanisms for precise DC performance prediction. We rigorously evaluate our model against leading architectures – long short-term memory (LSTM), attention-based (att-LSTM), convolutional LSTM (CNN-LSTM), gated recurrent unit (GRU), and CNN-GRU – to affirm its superiority in predictive accuracy and robustness. The integration of convolutional layers processes hourly data inputs efficiently, reducing complexity and improving pattern detection. A subsequent flattening layer optimizes accuracy, while a dual-layered LSTM and a deep neural network delve into frequency, temporal dynamics, and complex data relationships. Incorporating an attention mechanism into the att-CLDNN model has revolutionized predictive analytics in DC energy management, significantly enhancing accuracy by highlighting crucial data interdependencies. This model's unparalleled precision, evidenced by achieving the lowest Mean Squared Error (MSE) of 0.000179, the minimum Mean Absolute Error (MAE) of 0.01048, and the highest R2 Score of 0.977031, underscores its effectiveness. Crucially, this breakthrough fosters sustainability in energy management, promoting greener DC operations through precise energy use predictions, leading to substantial energy savings and reduced carbon emissions, in alignment with global sustainability objectives.

Social media has swiftly established itself as a primary source of products’ information for customers. Nowadays, Small and Medium-size Enterprises (SMEs) can use social media to develop Customer Relationship Management platforms (social CRM). Firms, particularly SMEs in developing countries need to understand the factors affecting their performance by implementing social CRM. However, there is a dearth of awareness on the impact of social CRM on the performance of SMEs. This study proposes an integrated model that aims to investigate the effects of social CRM on SMEs’ performance. The model is constructed by incorporating three dominant theoretical frameworks: the Fit-Viability Model (FVM), Network Externalities, and the Resource-Based View (RBV). A cross-sectional survey was used to gather data from 149 SMEs managerial staff. Findings revealed that almost 50% of the variability in the performance of SMEs is explained by the fitness and viability of social CRM. In addition, network externalities of social media significantly impact the social CRM fitness in the context of SMEs with path coefficient 0.617. Furthermore, the internal financial resources factor makes sCRM viable for SMEs as the results show significant relationship between the internal financial resources and the sCRM viability with 0.536 path coefficient. Manager innovativeness, IT knowledge, top management support, and government assistance, on the other hand, do not contribute significantly to the viability of social CRM for SMEs. The model aids SMEs in making well-informed decisions regarding the adoption of social CRM by evaluating both the suitability of social media for CRM tasks and the enterprise preparedness to implement social CRM, leading to enhanced performance.

Abstract The calibration of reservoir models using production data can enhance the reliability of predictions. However, history matching often leads to only a few matched models, and the original geological interpretation is not always preserved. Therefore, there is a need for stochastic methodologies for history matching. The Ensemble Kalman Filter (EnKF) is a well-known Monte Carlo method that updates reservoir models in real time. When new production data becomes available, the ensemble of models is updated accordingly. The initial ensemble is created using the prior model, and the posterior probability function is sampled through a series of updates. In this study, EnKF was employed to evaluate the uncertainty of production forecasts for a specific development plan and to match historical data to a real field reservoir model. This study represents the first attempt to combine EnKF with an integrated model that includes a genuine oil reservoir, actual production wells, a surface choke, a surface pipeline, a separator, and a PID pressure controller. The research optimized a real integrated production system, considering the constraint that there should be no slug flow at the inlet of the separator. The objective function was to maximize the net present value (NPV). Geological data was used to model uncertainty using Sequential Gaussian Simulation. Porosity scenarios were generated, and conditioning the porosity to well data yielded improved results. Ensembles were employed to balance accuracy and efficiency, demonstrating a reduction in porosity uncertainty due to production data. This study revealed that utilizing a PID pressure controller for the production separator can enhance oil production by 59% over 20 years, resulting in the generation of 2.97 million barrels of surplus oil in the field and significant economic gains.

An intelligent prediction model for rice yield in small-scale cultivation areas can provide precise forecasting results for farmers, rice planting enterprises, and researchers, holding significant importance for agricultural industries and crop science research within small regions. Although machine learning can handle complex nonlinear problems to enhance prediction accuracy, further improvements in models are still needed to accurately predict rice yields in small areas facing complex planting environments, thereby enhancing model performance. This study employs four rice phenotypic traits, namely, panicle angle, panicle length, total branch length, and grain number, along with seven machine learning methods—multiple linear regression, support vector machine, MLP, random forest, GBR, XGBoost, and LightGBM—to construct a yield prediction model group. Subsequently, the top three models with the best performance in individual model predictions are integrated using voting and stacking ensemble methods to obtain the optimal integrated model. Finally, the impact of different rice phenotypic traits on the performance of the stacked ensemble model is explored. Experimental results indicate that the random forest model performs best after individual machine learning modeling, with RMSE, R2, and MAPE values of 0.2777, 0.9062, and 17.04%, respectively. After model integration, Stacking–3m demonstrates the best performance, with RMSE, R2, and MAPE values of 0.2483, 0.9250, and 6.90%, respectively. Compared to the performance after random forest modeling, the RMSE decreased by 10.58%, R2 increased by 1.88%, and MAPE decreased by 0.76%, indicating improved model performance after stacking ensemble. The Stacking–3m model, which demonstrated the best comprehensive evaluation metrics, was selected for model validation, and the validation results were satisfactory, with MAE, R2, and MAPE values of 8.3384, 0.9285, and 0.2689, respectively. The above research findings demonstrate that this integrated model possesses high practical value and fills a gap in precise yield prediction for small-scale rice cultivation in the Yunnan Plateau region.

Estimating species distribution and abundance is foundational to effective management and conservation.Using an integrated species distribution model that combines presence‐only data from various sources with detection/non‐detection data from structured surveys, we estimated the distribution and expected abundance of three difficult‐to‐monitor mammals of management concern across New York State, namely, coyotes (Canis latrans), bobcats (Lynx rufus) and black bears (Ursus americanus). Three distinct landscape‐scale camera trap surveys provided detection/non‐detection data over 9 years between 2013 and 2021, and we augmented those data with incidental records of our focal species from public repositories. We used an inhomogeneous Poisson point process to construct an integrated model that fit both data types simultaneously.We demonstrate a simple application of spatial point density of all species records in the accessed public databases to inform the thinning process to account for unknown spatial sampling in the presence‐only data, often referred to as the 'magic covariate'. Using this approach, we examine habitat associations and provide spatially explicit estimates in expected abundance across the entirety of New York State for all three focal species.As expected, coyotes were the most widely distributed and abundant species, with a strong positive association with agricultural land uses. Bobcats exhibited low expected abundance throughout the state and showed positive associations with deciduous forest and forest edge, and a negative association with road density. Finally, we observed considerable spatial variation in abundance of black bears with expected abundance increasing in association with various forest cover and composition covariates and decreasing with crop cover. We present insights into habitat associations and spatial variation in abundance, and provide management implications for each of the species of interest.Synthesis and applications. Our integrated modelling method allows for managers to use citizen sightings combined with detection/non‐detection surveys to estimate robust indices of abundance for both high‐ and low‐density, and wide‐spread versus patchily distributed species. Through comparison with previous studies, we highlight how broad‐scale programmes, such as the statewide efforts to estimate species distributions undertaken here, can benefit substantively from integrated models that leverage additional data (here, incidental records) from a larger region of space, and thus capture more landscape heterogeneity than is plausible within formalized surveys alone. [ABSTRACT FROM AUTHOR]

Disasters caused by heavy rainfalls are of growing concern to researchers and government officials. While many studies have provided details of rainstorm-induced risks and efficient strategies for stormwater management, there is still a lack of attention to how the interactions between urban sewer systems and road networks during precipitation events affect sewer system performance and road inundation. To fill this gap, we have developed an integrated model that combines hydraulic characteristics and the topological structure of a sewer-road network system to explore the behaviour of these two interdependent systems and identify risk components during precipitation events. We apply the model to a watershed during different return periods of precipitation events in Cincinnati, Ohio, USA. The results reveal that the behaviour of some inconspicuous pipes has a significant impact on the sewer-road network system, resulting in a significant decrease in the system performance. Moreover, the interactions between road and sewer networks create multiple microstructures of connected components, which leads to different risks of interdependent systems and road inundations. The modelling results provide target areas for mitigation projects to reduce rainstorm-induced risks. [ABSTRACT FROM AUTHOR]