Search Results (231)

In bank-dominated financial systems, the accumulation of non-performing loans is a recognised source of systemic vulnerability, as correlated corporate defaults can erode bank capital, impair liquidity, and propagate stress across interconnected portfolios. Firm-level default detection thus constitutes a microprudential foundation of macroprudential stability: the reliable early identification of risky borrowers reduces both individual credit losses and the aggregate exposures that drive system-level fragility. Yet the use of structured data-mining pipelines for this task remains underexplored in Central and Eastern Europe. This study applies the CRISP-DM methodology to predict corporate loan default using data on 302 Slovak corporate borrowers, combining financial ratios from publicly available financial statements with selected company and loan-related information from internal bank records. Seven individual classifiers were developed and compared: decision trees (CART, CHAID, C5.0), logistic regression, discriminant analysis, and neural networks (MLP, RBF), together with a stacked ensemble based on their outputs. Model performance was evaluated using sensitivity, overall classification accuracy, and area under the ROC curve (AUC), with sensitivity treated as the primary criterion because of the asymmetric costs of misclassification in credit risk assessment. The results confirm that historical firm-level information provides a reliable basis for default prediction, with tree-based models consistently outperforming statistical and neural network approaches. The stacked ensemble achieved the strongest overall performance, whereas C5.0 and CHAID showed that interpretable classifiers can also deliver competitive predictive accuracy. A champion–challenger deployment architecture is proposed, in which the ensemble serves as the performance-oriented champion and interpretable models act as challengers; this arrangement contributes to the operational resilience of the credit-risk assessment process and aligns with macroprudential expectations of model governance, auditability, and explainability. The study offers a replicable methodological framework for integrating data-driven decision support into credit evaluation in comparable banking settings. Full article

The Copenhagen 10-step method is a set of policies that originated in the 1950s to reduce vehicle traffic in Copenhagen, which was heavily impacted by traffic. These policies are incorporated into a different dynamic on a global scale every day and are adopted while maintaining relevance. These policies, advocated in the context of climate change and carbon emission targets, as well as livability and health-focused urbanization, justice, and accessibility in transportation, are criticized for potentially negatively affecting low-income groups and commercializing urban transformation. Furthermore, they require adaptation because their applicability is seen as limited in terms of localization. In this context, the adaptability of the method to different social and spatial contexts has become a critical research topic, particularly in local studies, where application is more important and the order of implementation becomes of great importance. Within the scope of this study, a Copenhagen 10-step prioritization study was conducted specifically for the Küçükçekmece Kanarya Neighborhood, where low-to-middle socioeconomic groups live, and which has been declared a risky area in terms of building stock. Accordingly, a two-phase study was conducted. In the first phase, transportation and planning experts were asked to prioritize the 10 steps, and the timing of each implementation was determined based on the resulting ranking. In the second phase, accessibility analyses for the region were conducted using GIS (Geographical Information Systems)-based spatial data, such as accessibility, slope, and the distribution of urban facilities. Subsequently, these two phases were combined to create a simple prioritization framework for the areas of greatest concern in Kanarya, as well as for urban renewal, transportation, and government investment plans. According to the SWARA results, increasing bicycle use (C10) was the most important criterion at 17.2%, followed by making the bicycle the primary mode of transportation (C9) at 13.8% and adapting the city to seasonal changes (C8) at 11.5%. This study, which is significant for its focus on a specific region at the local implementation level, presents a straightforward model—based on concrete findings—for prioritizing sustainable transportation and urbanization policies in socioeconomically vulnerable areas. In doing so, it contributes to aligning theoretical approaches with practical field applications. Full article

Cryptosporidium spp. is a widely distributed gastrointestinal pathogen in vertebrates, such as the European wild boar. Furthermore, with a fecal–oral pathway, they might spread through tainted food and water or by direct contact. Related to the presence of this parasite in wild boar populations, the handling of hunted carcasses may be a source of zoonotic transmission. This work aims to evaluate the presence of Cryptosporidium spp. in 10 Portuguese hunting areas in two different locations (Northern and Central Portugal) and to preliminarily assess the risk factors of zoonotic transmission to hunting stakeholders. Cryptosporidium spp. antigens were confirmed by an immunochromatography test in the wild boars’ fecal samples from four of the 10 hunting areas analyzed (one in the North and three in the Southeast of Central Portugal). A qualitative assessment of various potential factors contributing to the persistence of infection in this wild population, but also of zoonotic risk factors related to hygiene procedures and handling of carcasses after hunting actions, was also carried out. With these potentially risky practices, it is imperative to raise awareness and establish a surveillance network in the hunting areas in order to mitigate the potential zoonotic transmission of these pathogenic agents to hunting stakeholders. Full article

Despite extensive research on traffic safety, the complex, non-linear spatial discrepancy between risky driving and actual accidents remains a significant challenge to quantify within diverse urban contexts. This study investigates the non-linear relationship between grid-level risky driving patterns and traffic accident occurrence in Daejeon, Korea, examining how these associations vary across different urban contexts. Using data collected from July 2023 to June 2024, the analysis incorporates GPS-based risky driving indicators, including rapid acceleration, deceleration, and sudden maneuvers from general passenger vehicles, thereby overcoming the limitations of previous studies reliant on commercial vehicle data. By adopting an H3-based spatial grid system, the study classifies areas into four quadrants based on median values of risky behaviors and accident counts, further categorizing them into “Matched” and “Mismatched” types to identify spatial discrepancies. Furthermore, the Explainable Artificial Intelligence (XAI) technique is employed to integrate regional variables—including population density, land use, and transport infrastructure—to uncover the key drivers of accident risks. Providing a significant methodological improvement over traditional linear models, the findings demonstrate that identical driving behaviors can yield different safety outcomes depending on local environmental interactions. Specifically, while driver behavioral factors directly explain accident frequency in matched regions, accident risks in mismatched regions are more significantly shaped by spatial environmental factors, such as green spaces and commercial land use, which override direct behavioral impacts. This study provides a robust framework for developing data-driven, region-specific traffic intervention strategies, including context-aware advanced driver assistance systems (ADAS) and spatially tailored traffic calming, to enhance urban safety. Full article

Organizations are currently in a stage where the volume of financial transactions and data is constantly growing. The same goes for risks associated with the use of data for risk management and strategic decision-making. The likelihood of transactional errors generally increases with data volume and process complexity, while fraud, although less frequent, may have more severe financial, compliance, and reputational consequences for organizations. Continuous auditing practices and well-established enterprise risk management (ERM) processes, combined with AI-driven pattern recognition, trend analysis and segmentation, can enhance timely detection and proper investigation of suspicious transactions. In areas with large volumes of transactions, the audit sampling process may be a lengthy process and pose a detection risk. Using machine learning (ML) models to support critical business processes could prove effective in managing enterprise risk overall. The current study offers new perspectives on managing risk and assurance with ML model output for flagging possible risky transactions within ERP (SAP) systems data. The study population consists of 69,158 finalized billing records extracted from the SAP production environment of a private sector organization, which covers a six-month operational period. The dataset was divided into an 80/20 train–test split, yielding 55,326 training and 13,832 test instances across six classification categories. The study examines the ML methods’ outcomes from billing datasets and their applicability in enhancing audit, assurance, and ERM processes by evaluating output data results from two supervised classification algorithms—multinomial logistic regression (SoftMax regression) and XGBoost—against various criteria generally accepted as risky in audit engagements. Model performance was assessed using accuracy, precision, recall, F1-score, ROC-AUC, and average precision (AP) from precision–recall curves. The results confirm that XGBoost achieves 99% overall accuracy with a macro F1-score of 0.965, outperforming logistic regression (macro F1 = 0.863), and that ML output allows early investigation and follow-up procedures to minimize the risk of fraud and errors and optimize risk management activities, thus strengthening internal control frameworks. Full article

Artificial intelligence and machine learning techniques can be used for performing magnetic testing on spacecraft that has historically been difficult and risky to perform. Some of the difficulty arises from the need to take these measurements from within the turbulent near-field area of the spacecraft. Some methods of testing require the spacecraft to be hoisted in the air and swung while the measurements are being taken so that any magnetic signatures in the test area can be removed. These new artificial intelligence and machine learning techniques can be used to determine the magnetic torque of complex magnetic systems. Here we will describe a test method that collects such data and poses much less risk to the spacecraft. We will also show some combinations of hyper-parameters that can be used to increase the speed and accuracy of the models. Some models were able to achieve over 96.6% accuracy of multipole determination on simulated data and over a 99.99% accuracy of dipole moment determination on simulated data. Applications include attitude control systems (ACS), science instrument locations, and data analysis. Full article

ERP systems significantly optimize many business processes and activities, but often their implementation and use in companies is a risky endeavor. They are the subject of various scientific studies and analyses in the fields of business, accounting and finance. The main focus in them falls on the process of implementing these systems, while the subsequent stages, risk analysis and long-term strategy are less affected. On this basis, this research paper proposes a methodology for quantitative assessment of identified ERP risks. It is based on a five-level matrix measuring three risk factors—influence, impact and vulnerability. The methodology has been empirically tested in three companies, different in size and operating in different economic sectors. The results show that the level of risk depends not only on the scale and complexity of the business, but also on the degree of integration of ERP solutions. Periodic application of the risk assessment methodology helps identify problem areas and facilitates management decision-making. Full article

Background: Sexually transmitted infections (STIs) represent a significant public health problem due to their impact. Knowledge about them, perceptions of the risk of contracting them, and adherence to prevention strategies such as HPV vaccination are, at various levels, key factors in preventing the spread of STIs. The study therefore aimed to investigate and evaluate, in a group of young Italians, the level of knowledge, perception of risk and propensity to adhere to preventive strategies, including vaccination against papillomavirus. Methods: A cross-sectional study was conducted by administering a questionnaire to young people aged between 16 and 30, residing in four macro-geographical areas, collecting socio-demographic, behavioral and knowledge data. Levels of knowledge about STIs and HPV were classified into four categories (low, medium without awareness, medium with awareness, high). Risk perception was assessed on a scale of 1 to 10. Results: A total of 2576 questionnaires were collected, revealing that general knowledge about STIs is limited: only 12.5% of participants demonstrated a high level of knowledge, while 27.1% demonstrated a low level; with regard to HPV, 41.3% of the sample demonstrated a low level of knowledge. The perception of the risk of contracting HIV and HPV was low in most subjects (average score of approximately 2.9 out of 10), with no significant differences related to levels of knowledge about HPV. Potential adherence to HPV vaccination was high (83.0% considered vaccination useful), but among unvaccinated subjects, almost half expressed concerns about vaccination, related to poor knowledge and mistrust of vaccines in general. Factors associated with a higher frequency of self-reported STIs included older age, transgender identity, non-heterosexual orientation, and risky sexual behavior. Conclusions: The results emerging from the study highlight the urgent need to strengthen educational and preventive interventions aimed at young people. Raising awareness of the risk of contracting STIs and the importance of vaccination are key targets for health promotion interventions. Full article

Background: Human papillomavirus (HPV) is the leading cause of sexually transmitted infections (STIs). It is found in extragenital regions, including the oropharyngeal cavity. Its presence in this area is linked to the increased prevalence of oral and pharyngeal cancer cases in young individuals, which is associated with current sexual practices in the young population. Objective, the objective of this study was to estimate the prevalence of HPV infection in the oropharyngeal cavity and identify associated factors within the student community of the Engineering and Chemical Sciences Unit of the University of Veracruz. Methods: an observational, descriptive, and transversal study was conducted. The study included 136 sexually active students aged 18 to 25 without oropharyngeal infection. After obtaining informed consent from all participants, mouthwashes were collected from the oropharyngeal cavity for subsequent detection of viral DNA and HPV genotyping using the PCR-RFLP technique. Risk factors were further assessed through a private questionnaire. For statistical analysis, a bivariate analysis of the main risk factors was performed, and Odds Ratios (OR) and 95% Confidence Intervals (CI) were calculated. Results: The results showed that HPV was detected in 6 participants, resulting in a prevalence of 4.4% (95% CI, 0.92–7.91), with genotypes 11, 52 and 58 identified. Notably, participants with a sexual orientation other than heterosexual had a 7.5-fold higher association with HPV. Conclusions: these findings indicate that low- and high-risk HPV infection in the oropharyngeal cavity is associated with risky sexual behavior in young individuals. Therefore, understanding the specifics of sexual activities is necessary to better comprehend viral transmission and spread among HPV-positive students. Full article

Corvids exhibit avoidance behaviour when foraging in the presence of potentially risky stimuli, yet it remains unclear how stimulus characteristics influence the strength of such responses. In this paper, we present wild-caught Large-billed Crows (Corvus macrorhynchos) with five conditions: no visual stimulus, a cardboard box (non-biological, stationary), an immobile owl decoy (biological, stationary), a continuous-motion owl decoy (biological, moving), and a sensor-activated-motion owl decoy (biological, moving, and sudden). Avoidance was quantified using feeding latency, landing frequency, total time spent in the feeding area, and food consumption. Compared with the condition with no visual stimulus, the presence of any visual stimulus elicited increased latency, indicating that crows detect and respond to objects near food. Among the four objects, the sensor-activated-motion owl decoy produced stronger avoidance responses of the crows than the non-biological and stationary object (cardboard box). This indicates that they evaluate not only the presence of an object but also its motion characteristics and/or perceived biological cues when adjusting their foraging behaviour. Although sample size and individual variation impose limitations, these findings suggest that both the presence of visual stimuli and/or the complexity of their appearance play key roles in shaping avoidance behaviour in corvids. Full article

Background: Pain after giving birth is commonly observed in horses, yet there has not been a specific tool developed for assessing this pain in postpartum mares. The goal was to adapt existing equine pain scales and to preliminarily validate a practical pain scale for use by veterinarians and caregivers after foaling. Methods: The pain scale was developed by adapting items from other pain scales, including established orthopedic and colic equine pain scales, and incorporating caregiver feedback. The final scale includes eight areas for assessing pain: behavior, facial expressions, vital signs, udder examination, gastrointestinal function, hoof temperature, response to food, and movement. Observations were conducted on ten heavy draft mares that experienced dystocia, with pain scores recorded twice daily for 1 to 4 days postpartum. Simultaneous saliva samples were collected to measure cortisol levels. Results: The pain scale proved feasible for use at the stall and allowed for partial scoring when certain assessments were deemed risky. Pain scores were highest on the first day after foaling and decreased as the mares recovered. In a case of clinical deterioration, a substantial increase in pain score was noted. Increased pain scores were associated with elevated cortisol levels, supporting the biological relevance of the scale. In clinical practice, if a pain score exceeded 40% of the maximum score, the mare was identified as a patient requiring analgesic treatment. Conclusions: This postpartum-specific pain scale provides a standardized method for assessing pain in mares after foaling and may assist in guiding appropriate pain management. Although the proposed pain scale shows promise as a clinical tool, the present results are preliminary and require confirmation in larger studies. Full article

Elite capture, a power structure problem involving rent-seeking, hinders sustainable water resources management. Governments play crucial roles in instilling public legitimacy in water governance, a common-pool resource that benefits from cooperative solutions such as pilot competitions, co-monitoring, and inter-agency coordination. A study of South-to-North Water Diversion Projects in China showed how, when governments outsource small projects to local sub-contractors, a method named moderate supervision (ruo jiandu) can enable effective oversight, which is superior to a bidding model with strict supervision (qiang jiandu). The concept of moderate supervision was initiated in 2023, before which most small projects had been left in a risky state with no supervision (ling jiandu). Analysis of a case in Shandong Yellow River Water Diversion Irrigation Area involved semi-structured in-depth interviews. Findings revealed that an elite-government-villagers tripartite spiral was composed of 3 dimensions reshaping a positive elite culture: first, a whitelist of qualified local contractors; second, co-monitoring of multiple stakeholders with influence exerted by a three-tier mobilization system; third, inter-agency coordination innovatively enabling smooth functioning between policy entrepreneurs of formal institutions and local social governance of informal ones. Policy implications to underscore real-world applicability are provided. Full article

Background/Objectives: Antibiotic resistance represents one of the major global health emergencies, driven by the inappropriate use of antibiotics and persistent misconceptions among adults attending general medical clinics. This study, conducted on 325 participants recruited from general medical clinics in the province of Salerno, aimed to assess their knowledge, perceptions, and behaviors regarding antibiotic use. Methods: A cross-sectional, quantitative observational survey was conducted using a structured questionnaire based on the WHO tool and adapted to the local context. Results: The results show that the majority of participants take antibiotics only when prescribed by a doctor (90.2%), but risky practices such as self-medication (10%) and early discontinuation of therapy (16%) persist. In addition, 72% of subjects demonstrate incomplete knowledge about the independent management of drugs, and 86% mistakenly believe that resistance is limited to the individual rather than the community. The descriptive analysis stratified by age showed higher levels of awareness among subjects under 30 years of age, compared to significant knowledge gaps and inappropriate behaviors in the over-65 age group. Conclusions: Despite a good awareness of the need for medical prescriptions and the collective importance of the phenomenon, there are still critical areas of knowledge and incorrect practices that can promote the spread of antibiotic resistance. The data collected underscore the urgency of targeted educational strategies differentiated by age group, integrated with multi-channel communication interventions, in order to promote the appropriate use of antibiotics and contain the impact of one of the most serious global health emergencies. Full article

Using big data for risk analysis of construction projects is a largely unexplored area. In this traditional industry, risk identification is often based either on so-called domain expert knowledge, in other words on experience, or on different statistical and quantitative analysis of individual past projects. The motivation of this research is based on the implemented and evaluated data-driven and AI-based DARIA approach to identify financial risks in the execution phase of transportation infrastructure construction projects that shows exceptional results at an early stage of the project execution phase and has already been deployed into enterprise-wide production within the STRABAG group. Due to DARIA’s productive use, concern and doubts about the trustworthiness of its ML algorithm are certainly possible, especially when DARIA identifies risky projects while all conventional metrics within the STRABAG controlling system do not identify any problems. “If AI systems do not prove to be worthy of trust, their widespread acceptance and adoption will be hindered, and the potentially vast societal and economic benefits will not be fully realized”. Thus, and based on the results of a user study during DARIA’s successful deployment into enterprise-wide production, this paper focuses on the identification of suitable indicators to measure the trustworthiness of the DARIA ML algorithm in the interaction between individuals and systems as well as on the modeling of the reproducibility of the internal state of DARIA’s ML model. Full article

A shift in the public health goal for onchocerciasis from control to elimination implies that the treatment of onchocerciasis must be extended to communities that are hypoendemic for the disease. However, in such communities, the majority of the population may not manifest the symptoms of onchocerciasis. As a result, they may be reluctant to take part in epidemiological surveys aimed at monitoring parasite transmission, particularly due to the invasive nature of the currently approved diagnostic tests. This reluctance is compounded by the absence of visible, severe manifestations of the disease in these areas. On the other hand, diagnostic methods that utilize samples collected by a non-invasive procedure, such as urine, are generally painless and not risky. In this context, we evaluated the diagnostic performances of OvMANE1 and OvMCBL02 multiepitope antigens using urine samples. The evaluation of total IgG and IgG subclass responses revealed IgG3 as the most effective IgG for the OvMANE1 test (sensitivity = 87.5%, specificity = 100.0%), total IgG for the OvMCBL02 test (sensitivity = 92.5%, specificity = 100.0%), and IgG3 for the OvMANE1_OvMCBL02 cocktail test (sensitivity = 92.5%, specificity = 100.0%). These tests have the potential to meet the criteria of a diagnostic test’s target product profile to map onchocerciasis in low-prevalence areas, where a sensitivity of ≥60.0% and specificity of ≥99.8% are recommended. Furthermore, the OvMCBL02 and OvMANE1_OvMCBL02 cocktail tests may have the features of a diagnostic test’s target product profile to determine treatment endpoints (recommended sensitivity ≥ 89.0%, specificity ≥ 99.8%) as reported by the Diagnostics Technical Advisory Group for Neglected Tropical Diseases of the World Health Organization. Consequently, further characterization of these multiepitope antigens may enable urine, which can be collected non-invasively, to be used in the OvMANE1 and OvMCBL02 tests for the field evaluation of onchocerciasis. Full article