Test Of Model Research Articles

Estimation of barrier layer thickness from satellite-derived sea surface parameters in the north Indian Ocean using machine learning

ABSTRACT Oceanic barrier layer thickness (BLT) has been estimated from sea surface parameters through four machine learning (ML) models. The BLT used for training and testing of ML models has been computed from 19 years (2005–2023) of Argo data of vertical temperature and salinity profile in the north Indian ocean. It is observed that thick barrier layer (BL) exists in three regions within the north Indian ocean: the Bay of Bengal (BoB), Eastern Arabian Sea (EAS) and the East Equatorial Indian Ocean (EEIO) off the Sumatra. The observed BLT ranges from 0 to 80 m in the north Indian ocean with a seasonal cycle. Four ML models, i.e. multiple linear regression, artificial neural network (ANN), Random Forest and XGBoost have been trained for 2010–2018 to learn the relationship between sea surface parameters (input features) and BLT (target/output variable) in three thick BL regions within the north Indian Ocean. Thereafter, the trained models have been tested for the estimation of BLT in the same regions for the time period of 2019–2023. The ML models are well able to estimate the BLT from sea surface parameters (SST, SSS) with RMSE less than 5 m in the BoB and EAS, which is less than 8 percentile of the model testing datasets. The ‘Random Forest Regressor’ turns out to perform better than the other three ML models for estimation of BLT. Further, the Random Forest estimated BLT has been compared with the BLT computed using ORAS5 reanalysis data of vertical temperature and salinity profile. This study explores the capability of the ML models for capturing the signature of a subsurface variable on the surface owing to the limitation of costly subsurface measurement.

Soil moisture retrieval over agricultural region through machine learning and sentinel 1 observations

Soil moisture is a fundamental variable in the Earth’s hydrological cycle and vital for development of agricultural water management practices. The present study provided a comprehensive evaluation of a wide range of advanced machine learning algorithms for Soil Moisture (SM) estimation from microwave Synthetic Aperture Radar (SAR) backscatter observations over the wheat fields. From the wheat fields, samplings were performed to collect the in situ datasets on three different dates concurrent to the Sentinel-1 overpasses. The backscattering coefficients were taken as the input variables and SM as the output variable for the training and testing of different models. The performance analysis of RMSE, R-squared, and correlation coefficients revealed that the Random Forest (RF) and Convolutional Neural Network (CNN) models demonstrated superior performance for SM estimation over the wheat field. Specifically, the RF model exhibited outstanding accuracy and robustness in both the training [RMSE (%): 3.44, R-squared: 0.88, correlation: 0.95] and validation phases [RMSE (%): 7.06, R-squared: 0.61, correlation: 0.8], marking it as the most effective model followed by the CNN model with [RMSE (%): 3.9, R-squared: 0.84, correlation: 0.92] during training and [RMSE (%): 8.44, R-squared: 0.43, correlation: 0.67] for validation, highlighting challenges in the model generalisation.

Reporting of Regression Models for Ordinal Responses in Sports Sciences Field: A Systematic Review

ABSTRACTThe integration of statistical methods in sports sciences has become essential for decision‐making in performance analysis, injury prevention, and athlete outcomes. This work presents a systematic review following PRISMA guidelines to explore the application of ordinal regression models in the sports sciences field. A comprehensive search of articles published prior to March 4, 2023 identified 34 included studies. This search included widely recognized databases such as Web of Science, PubMed, and specialized journals of sports statistics, such as the Journal of Quantitative Analysis in Sports, and the Journal of Sports Analytics. The analysis reveals that 26.5% of these articles were published in statistics and sports statistics journals. Notably, R emerged as the primary software used for analysis in 38.8% of the studies. However, a significant majority (82.4%) of the studies did not provide data and code repositories. Among sports, soccer had the highest representation (28.6%), followed by basketball (17.1%). The most reported ordinal model was the proportional odds model (32.3%), followed by the mixed effects proportional odds model (11.8%), while a relevant proportion (29.4%) did not report the model used. Furthermore, 23.5% of articles proposed novel models. Validation tests for proportional odds models were not conducted in 53.3% of cases. This review underscores the importance of improved reporting practices, inclusivity in sport representation, and statistical education in advancing sports analytics.

Selection and evaluation of a set of attributes appropriate for detection of antisocial behaviour in online media

Abstract Nowadays the world of modern technologies brings new ways of communication and interaction between people. Online communication becomes faster and more convenient, but it also enables an exchange of more dangerous information. Antisocial behaviour in online web discussions becomes one of the most serious problems. This paper is focused on the analysis and identification of the most typical attributes of antisocial behaviour in the online space. Our research attempts to distinguish the most characteristic features of suspicious contributors to identify attributes that can define antisocial behaviour in the best way. The main objective is to evaluate the success of these attributes in the automatic detection of the suspicious contributors—trolls using classification methods of machine learning such as naïve Bayes, decision trees, random forest, logistic regression, and support vector machine. The methods were selected from the point of view of evaluation of suitability of individual attributes and evaluation of selected sets of attributes. The results of test of models learned using mentioned methods are discussed from this point of view and lead to a selection of a small set of attributes from all considered attributes.

A test of the Grant-Stebbins pollinator-shift model of floral evolution.

Pollinators are thought to play a key role in driving incipient speciation within the angiosperms. However, the mechanisms underlying floral divergence in plants with generalist pollination systems, remains understudied. Brunsvigia gregaria displays significant geographical variation in floral traits and are visited by diverse pollinator communities. Because pollinators are often shared between populations, we investigated whether specific pollinators are responsible for driving floral divergence between them. Three distinct ecotypes were identified, each dominated by three different pollinators: bees, swallowtail butterflies, and long-proboscid flies. Across seven populations, we found a pattern of association between style length and the morphology of pollinators that visit the flowers most frequently and contact the reproductive parts most often. Furthermore, we found significant linear, quadratic and correlational selection on flower number, tepal length and style length within the butterfly- and bee-dominated populations. We also found partial evidence for divergent selection on these traits between experimental sites. Our findings suggest that a handful of key pollinators that vary in their importance have the potential to drive population-level divergence in floral traits, which may lead to pollination ecotype formation.

Seismic anisotropy prediction using ML methods: A case study on an offshore carbonate oilfield.

Estimating seismic anisotropy parameters, such as Thomson's parameters, is crucial for investigating fractured and finely layered geological media. However, many inversion methods rely on complex physical models with initial assumptions, leading to non-reproducible estimates and subjective fracture interpretation. To address these limitations, this study utilizes machine learning methods: support vector regression, extreme gradient boost, multi-layer perceptron, and a convolutional neural network. The abundance of seismic features leads to many feature combinations, making the training and testing of machine learning models challenging. Therefore, a workflow has been developed to systematically inspect seismic features and select the most appropriate one for anisotropy estimation with reasonable accuracy. Synthetic data were generated using an earth model and well data within a finite difference numerical program. After thoroughly investigating synthetic data, the amplitudes of direct and reflected waves in the time and frequency domains were selected as input features to train machine learning methods. Optimizing the machine learning hyperparameters allowed the training and testing procedures to be completed with high accuracy. Subsequently, the optimized machine learning methods were used to predict Thomsen's parameters, ε and δ, of a shaley formation in the zone area. To validate the predictions, the ε and δ estimated at a well location were compared with those obtained using a physics-based model, resulting in the least relative errors ranging from 2.92% to 7.14%.

Galleria mellonella (Greater Wax Moth) as a Reliable Animal Model to Study the Efficacy of Nanomaterials in Fighting Pathogens.

The spread of multidrug-resistant microbes has made it necessary and urgent to develop new strategies to deal with the infections they cause. Some of these are based on nanotechnology, which has revolutionized many fields in medicine. Evaluating the safety and efficacy of these new antimicrobial strategies requires testing in animal models before being tested in clinical trials. In this context, Galleria mellonella could represent a valid alternative to traditional mammalian and non-mammalian animal models, due to its low cost, ease of handling, and valuable biological properties to investigate host-pathogen interactions. The purpose of this review is to provide an updated overview of the literature concerning the use of G. mellonella larvae as an animal model to evaluate safety and efficacy of nanoparticles and nanomaterials, particularly, of those that are used or are under investigation to combat microbial pathogens.

Research on robustness testing and enhancement method for deep learning models

Research on robustness testing and enhancement method for deep learning models

Maximum Conditional Alpha Test for Conditional Multi-Factor Models

Maximum Conditional Alpha Test for Conditional Multi-Factor Models

Comparison of cardiovascular risk prediction models developed using machine learning based on data from a Sri Lankan cohort with World Health Organization risk charts for predicting cardiovascular risk among Sri Lankans: a cohort study

IntroductionModels derived from non-Sri Lankan cohorts are used for cardiovascular (CV) risk stratification of Sri Lankans.ObjectiveTo develop a CV risk prediction model using machine learning (ML) based on data from...

Deep learning can detect elbow disease in dogs screened for elbow dysplasia.

Medical image analysis based on deep learning is a rapidly advancing field in veterinary diagnostics. The aim of this retrospective diagnostic accuracy study was to develop and assess a convolutional neural network (CNN, EfficientNet) to evaluate elbow radiographs from dogs screened for elbow dysplasia. An auto-cropping tool based on the deep learning model RetinaNet was developed for radiograph preprocessing to crop the radiographs to the region of interest around the elbow joint. A total of 7229 radiographs with corresponding International Elbow Working Group scoring were included for training (n=4000), validation (n=1000), and testing (n=2229) of CNN models for elbow diagnostics. The radiographs were classified in a binary manner as normal (negative class) or abnormal (positive class), where abnormal radiographs had various severities of osteoarthrosis and/or visible primary elbow dysplasia lesions. Explainable artificial intelligence analysis were performed on both correctly and incorrectly classified radiographs using VarGrad heatmaps to visualize regions of importance for the CNN model's predictions. The highest-performing CNN model showed excellent test accuracy, sensitivity, and specificity, all achieving a value of 0.98. Explainability analysis showed frequent highlighting along the margins of the anconeal process of both correctly and incorrectly classified radiographs. Uncertainty estimation using entropy to characterize the uncertainty of the model predictions showed that radiographs with ambiguous predictions could be flagged for human evaluation. Our study demonstrates robust performance of CNNs for detecting abnormal elbow joints in dogs screened for elbow dysplasia.

Poincaré plot analysis of ECG uncovers beneficial effects of omaveloxolone in a mouse model of Friedreich's ataxia.

Friedreich's ataxia (FA) is a rare inherited neuromuscular disorder, where most patients die from lethal cardiomyopathy and arrhythmias. Mechanisms leading to arrhythmic events in FA patients are poorly understood. This study aims to examine cardiac electrical signal propagation in mouse model of FA with severe cardiomyopathy and evaluate effects of omaveloxolone (OMAV), the first FDA-approved therapy. Cardiac-specific MCK-Cre frataxin knockout (FXN-cKO) mice were used to mimic FA cardiomyopathy. In vivo surface ECG recordings, western blotting, qPCR, and histochemistry were performed. Characteristics like long QT syndrome, interatrial block, and ST-segment abnormalities in FA patients were identified in FXN-cKO mice. FXN-cKO mice exhibited sexual dimorphism in electrical signal propagation and cardiac structural integrity. Untreated FA males showed increased ventricular propagation intervals, while females exhibited delayed atrial propagation. OMAV showed no significant therapeutic effect on average ECG time intervals but improved chamber-specific waveforms when aggregated frequency distributions were analyzed. The J-wave was absent in FXN-cKO male mice but reappeared with OMAV treatment. Poincaré plots revealed disparate idiopathic arrhythmias with multi-clustering events in individual mice with high incidence in FXN-cKO males. OMAV treatment reduced multi-clustering events to a single cluster; however, ANS dysfunction still remained. Our study revealed significant electrical propagation disturbances and sexual dimorphism in FXN-cKO mice with severe cardiomyopathy. Poincaré plots identified irregularities in heart rhythm and ANS dysfunction. OMAV improved heart function by stabilizing early repolarization and reducing disparate arrhythmias. This work stresses sex-specific ECG interpretations and alternative mathematical approaches for drug testing in FA models.

HEC-RAS simulation of scour in circular piers using a sediment transport demonstration channel

Intending to take advantage of the implementation of two very affordable resources, a sediment transport demonstration channel for teaching and a very versatile open-source software (HECRAS), the present work develops the numerical simulation of the scour in a circular pier. The numerical modeling (HEC-RAS) was calibrated with the tests of the physical model (sediment transport demonstration channel). The reduced dimensions of the channel limited the number and precision of the tests; however, they were sufficient to provide the information required by the numerical model. The work details the physical and numerical modeling in such a way that a clear and simple guide is obtained, to simulate the scour of piers in bridges. The result obtained in the first instance exceeds the scour of the physical model, however, due to the simulated conditions, very close to a wide pier in shallow flow, the overestimation was expected.

Prevalence and Determinants of Depression Among University Students: Case Study, Technical University of Mombasa

Depression is a significant global mental health issue that affects countless individuals annually and can lead to severe consequences, including suicide. One area that has seen an increase in mental health concerns is among public university students, with many struggling with depression-related challenges and even dropping out of school. Unfortunately, there have been several reports of suicide among university students in the mainstream media. To better understand the situation, a study was conducted to determine the prevalence of depression among Technical University of Mombasa students and identify the risk factors that contribute to this condition. The study surveyed 390 students using structured questionnaires, with depression as the dependent variable and family, school, environment, and economic factors as the independent variables. Through the use of the chi-square test of independence and multiple linear regression model, the study found that 57.7% of students experienced depression, with students in year four and above being more likely to be affected than those in lower years. The study found that family, school, environment, and economic factors significantly influence depression among students, with 34.2% of the variations in depression explained by these factors. Family, school, environmental, and economic factors were found to increase depression levels by 23.8%, 12.7%, 10.6%, and 17.7%, respectively. Based on these results, the study recommended establishing peer counselling programs, raising awareness of mental health issues, creating online platforms for students to share their concerns, promoting participation in extracurricular activities, and incorporating mental health education into the university curriculum. These recommendations aim to address the rising cases of depression and promote the overall well-being of students in public universities in Kenya.

RECOGNIZING A TRANSFORMER OIL GAUGE IN AN IMAGE USING A PRE-TRAINED MOBILENETV2 FPN LITE MODEL

Oil gauges allow you to monitor the amount of oil in transformers, which is necessary to prevent its overheating and related negative consequences, such as deterioration of the insulating and cooling properties of the oil, increase in pressure inside the device and possible accidents. Most of the oil pointers do not have an electric output to indicate the measurement results, which complicates the process of reading the readings of the device. One of the options for solving this problem may be the use of intelligent methods of image analysis, which have recently been actively used in many scientific and practical fields. The purpose of the study was to develop and evaluate the effectiveness of an intelligent system for determining the presence of an arrow transformer oil gauge in an image using a pre-trained MobileNetV2 FPN Lite model. Materials and methods. In the course of the study, theoretical and empirical methods were used. The study considers the pointer oil gauges of the MS-1 and MS-2 models. To develop a machine learning model for image recognition, the Python language was chosen, the open-source TensorFlow Object Detection API platform, which is part of TensorFlow, was used, which implements the creation, training, and testing of neural network object detection models. Research results. The process of generating a training set using augmentation methods and the process of data labeling are described. The process of selecting hyperparameters for model training was considered and analyzed. The hyperparameter “number of neurons” was used with the same value as in the pre-trained model. ReLU was chosen as the activation function, and the momentum optimization method was used. The batch size was 4, and the number of epochs was assumed to be 2000. The Accuracy, Recall, Precision, and F1-score metrics were calculated, and the quality and efficiency of the resulting model were evaluated based on them. The model did a good job of avoiding false positives, which confirms the high value of the Precision = 1.0 metric. However, it incorrectly recognized a significant number of non-oil gauge objects, as reflected by the low Recall = 0.687 value. The reason for such indicators of the Recall and F1-score metrics is the imbalance of data for training. To eliminate this drawback in further research, it is necessary to prepare a larger number of unique images of this device, displaying it at different angles of inclination, in different weather conditions, in different lighting. Conclusions. A model of an intelligent system for determining the presence of a transformer oil gauge in an image using a pre-trained MobileNetV2 FPN Lite model has been built. The model was trained to determine the device and its boundaries on the image, but the accuracy of the model is insufficient for its use in real conditions. Increasing the training, validation and test sample size can solve this problem. In general, the proposed method of recognizing transformer oil indicator on the image has shown its performance.

Critical points in the views of G7’s country leaders on national agendas for SDGs and SFDRR

PurposeDisasters are catastrophic events that result in loss of life, injury and significant disruptions to everyday life. Unsustainable poverty, crises and migration policies worsen them. This study aimed to evaluate the critical perspectives of G7’s country leaders regarding their national agendas in 2022, focusing on the sustainable development goals (SDGs) and the Sendai Framework for Disaster Risk Reduction (SFDRR).Design/methodology/approachThis study utilized document analysis, one of the qualitative research methods. Using purposive sampling, we analyzed 20,215 words of X (Twitter) data from G7’s country leaders between January 1 and December 31, 2022, with a thematic approach using MAXQDA qualitative software. MAXQDA applied Code Cloud, Code Matrix Browser, Word Frequencies and MaxMaps Code Models tests. Results are presented based on codes, categories and themes identified during analysis.FindingsThe G7’s country leaders emphasized peace, justice and strong institutions as their SDGs priority. However, it was observed that developed nations often overlooked critical issues such as hunger, quality education and inequality in their agendas in 2022. Based on SFDRR, the leaders addressed crisis response and recovery topics, but it is crucial to note that there is a need for a shift towards proactive measures in their disaster policies. The study also noted limited coverage of migration issues affecting many countries globally. There is an urgent need to address hunger, provide quality education and reduce inequality worldwide.Research limitations/implicationsThis study has several limitations: Former Italian Prime Minister Mario Draghi did not have an X (Twitter) account in 2022, and the data only reflect the situation in 2022. Also, simply analyzing statements does not necessarily indicate what these countries support and promote in practice. However, this study could be enhanced by including analyses of funding allocations for projects that target various SDGs and SFDRR objectives. Lastly, this study focused on the strategic policy statements made by representatives of different countries. It does not include tactical and operational disaster management processes, which is a limitation because it is difficult to provide a broader perspective.Practical implicationsDeveloped countries’ approaches to disaster and development policies are essential for disaster risk reduction strategies. Future analyses of funding allocations for projects targeting various SDGs and SFDRR objectives can lead to practical implications.Originality/valueThe authors present a critical viewpoint on the SDGs and SFDRR perspectives of G7’s country leaders. As developed countries possess significant resources and capabilities, they should set a positive example in the global implementation of disaster policies. Consequently, their annual official statements are a valuable guide for disaster policymakers and practitioners.

Scientific, inquiry, and animation integration: The IPSIA learning model in chemistry

This study addresses the issue of low active participation among students in chemistry learning, as well as their difficulties in understanding macroscopic, microscopic, and symbolic concepts, which contribute to unsatisfactory learning outcomes. To tackle these challenges and enhance students' performance in chemistry, a valid, practical, and effective IPSIA learning model was developed. The research employed a development approach based on the ADDIE (Analyze, Design, Develop, Implement, and Evaluate) model, beginning with a thorough analysis of the curriculum. This was followed by the design of the IPSIA model, which was validated using a questionnaire instrument. The results of the validity test scored 98, categorizing it as "very good." Additionally, the practical test of the IPSIA model also received a score of 98, indicating that it is user-friendly. During the implementation phase, the effectiveness of the model was assessed through test results. The improvement in students' chemistry learning was evidenced by normalized gain scores: 0.57 for the control class and 0.72 for the experimental class at Senior High School (SMA) 2 Padangsidimpuan, as well as 0.52 for the control class and 0.64 for the experimental class at SMA 5 Padangsidimpuan. These findings demonstrate that the IPSIA learning model is valid, practical, and effective in improving students' chemistry learning outcomes.

Effects of Socioeconomic Status on the Health of Menopausal Mothers in Multicultural Families in Korea: A Test of the Reserve Capacity Model.

Based on the Reserve Capacity Model, this study examined the effect of SES(socioeconomic status) on health status on the health status of menopausal mothers of multicultural families, focusing on mediating effects of self-esteem, assimilation, integration, separation, marginalization, and daily life stress. This is a cross-sectional survey study with secondary analysis of national data on 683 perimenopausal mothers in multicultural families from the 9th Multicultural Adolescents Panel Study surveyed by the Korea Youth Policy Institute in 2019. The subjects were women aged 45-55years and foreign nationals by origin. For path analysis using observed variables, the Mplus 8.4 program was used. Self-esteem and integration were positively related to health status and negatively related to marginalization and daily life stress. Health status was higher when self-esteem and integration were high, and work-life stress and marginalization were low. To improve the perceived health status of menopausal women in multicultural families, rather than simply focusing on SES through support for employment status and income, it is necessary to improve self-esteem by resolving difficulties caused by SES and cultural adaptation to relieve stress in daily life and to provide multidimensional coping resources that can improve health by lowering risk.

Institutions and Economic Growth: A Panel Granger Causality Analysis

This paper revisits the direction of causality between institutions and economic growth for a sample of 119 countries over the period 1999-2018, divided into four groups according to income level: high income, upper middle income, lower middle income and low income. The study uses two institutional datasets, the International Country Risk Guide (ICRG) for the main estimation and the World Governance Indicators (WGI) for check the robustness of the results. Using the non-causality Granger test in a heterogeneous panel model with fixed coefficients, developed by Dumitrescu and Hurlin (2012), the empirical results show a unidirectional relationship for all panels except for lower middle-income countries, where causality is bidirectional. The findings also suggest that causality patterns are heterogeneous and depend on the level of development of the countries. Based on these results, we propose some interesting recommendations. The types of reforms to prioritize must be determined according to the direction of causality between institutions and economic growth. Moreover, heterogeneous causality implies the implementation of different policies adapted to the level of development of each panel, rather than considering a common policy.

Iterative Ratio Method: a method to map the depth to the Moho from gravity anomalies using linear approximate invariance

Interfaces characterized by a density contrast are common models to interpret gravity data. Two key parameters of the interface inversion are the density contrast and constant offset, which are free parameters. We propose to iteratively estimate the two free parameters by a new method: the iterative-ratio (ITRAT) method. The method is based on the linear approximate invariance of depth and free parameters and on the availability of several depth constraints. The free parameters are estimated iteratively to update the interface depth. By processing the synthetic Moho model, we find that the ITRAT method generally converges after two iterations with good accuracy for the free-parameter estimates and Moho depth. Random tests for the constant density model show that the accuracy of ITRAT method is slightly affected by the noise of gravity data and by the number of constraints, but it is sensitive to the uncertainty of constraints. For the variable density model, the ITRAT method will estimate a group of constant density contrast and constant offset to obtain a better Moho model, which has the lower depth error than the model with a mean density contrast. The new method is finally demonstrated by mapping the Moho of the Santos basin (Brazil) according to the depth constraints and regional gravity anomalies.