Search

Like many psychological scales, depression scales are ordinal in nature. Depression prediction from behavioural signals has so far been posed either as classification or regression problems. However, these naive approaches have fundamental issues because they are not focused on ranking, unlike ordinal regression, which is the most appropriate approach. Ordinal regression to date has comparatively few methods when compared with other branches in machine learning, and its usage is limited to specific research domains. Ordinal logistic regression (OLR) is one such method, which is an extension for ordinal data of the well-known logistic regression, but is not familiar in speech processing, affective computing or depression prediction. The primary aim of this study is to investigate proportionality structures and model selection for the design of ordinal regression systems within the logistic regression framework. A new greedy based algorithm for partial proportional odds model selection (GREP) is proposed that allows the parsimonious design of effective ordinal logistic regression models, which avoids an exhaustive search and outperforms model selection using the Brant test. Evaluations on the DAIC-WOZ and AViD depression corpora show that OLR models exploiting GREP can outperform two competitive baseline systems (GSR and CNN), in terms of both RMSE and Spearman correlation.

Monocular depth estimation, as one of the fundamental tasks of computer vision, plays a crucial role in three-dimensional (3D) scene understanding and perception. Usually, deep learning methods recover monocular depth maps using continuous regression manners by minimizing the errors between the ground-truth depth and the predicted depth. However, fine depth features may not be fully captured through layer-by-layer coding, which is prone to low spatial resolution depth maps and insufficient details. Furthermore, it usually converges slowly and suffers from unsatisfactory results. To tackle these issues, we propose a novel model, named context-based ordinal regression network (CORNet), to reconstruct monocular depth maps in the ordinal regression manner with context information in this paper. Firstly, we put forward a novel context-based encoder with a feature transformation (FT) module to learn context information and details from inputs, and output multi-scale feature maps. Then, we design a boundary enhancement module (BEM) with a spatial attention mechanism following each operation of feature fusion, which captures boundary features in the scene to enhance the border depth. Finally, a feature optimization module (FOM) is designed to fuse and optimize the multi-scale features and boundary features to strengthen depth learning. What’s more, we introduce an ordinal weighted inference to predict depth maps from probabilities and discretization values. Experiments and results on two challenging datasets, KITTI and NYU Depth V2, demonstrate that our proposed CORNet can estimate monocular depth maps effectively and obtain superior performance in capturing geometric features over existing methods.

We propose a novel Bayesian Ordinal Regression approach for multiple criteria choice and ranking problems. It employs an additive value function model to represent indirect Decision Maker’s (DM’s) preferences in the form of pairwise comparisons of reference alternatives. By defining a likelihood for the provided preference information and specifying a prior of the preference model, we apply the Bayesian rule to derive a posterior distribution over a set of all potential value functions, not necessarily compatible ones. This distribution emphasizes the potential differences in the abilities of these models to reconstruct the DM’s pairwise comparisons. Hence a distinctive character of our approach consists of characterizing the uncertainty in consequence of applying indirect preference information. We also employ a Markov Chain Monte Carlo algorithm, called the Metropolis-Hastings method, to summarize the posterior distribution of the value function model and quantify the outcomes of robustness analysis in the form of stochastic acceptability indices. The proposed approach’s performance is investigated in a thorough experimental study involving real-world and artificially generated datasets.

Many real world problems require the prediction of ordinal variables where the values are a set of categories with an ordering to them. However, in many of these cases the categorical nature of the ordinal data is not a desirable outcome. As such, regression models treat ordinal variables as continuous and do not bind their predictions to discrete categories. Prior research has found that these models are capable of learning useful information between the discrete levels of the ordinal labels they are trained on, but complex models may learn ordinal labels too closely, missing the information between levels. In this study, we use several datasets where the outcome is continuous and generate ordinal labels from this variable. The performance of two types of models, namely ordinal classification and continuous regression, is examined to determine the effect of model complexity. The experiment confirms previous findings that regression models trained on the synthetic ordinal labels reach optimal performance on the continuous outcome with less complexity than compared with performance on the ordinal labels. Additionally, the former overfit more quickly as complexity increases. This suggests that for machine learning settings where we would like to form a continuous analog for ordinal training labels, models should be trained with less complexity than what appears optimal given the observed performance on the ordinal labels reaches optimization after the underlying ground truth continuous measure’s optimization occurs.

Associations of environmental variables with physical activity and sedentary time using data from the Canadian Longitudinal Study on Aging, and the Canadian Urban Environmental Health Research Consortium (Canadian Active Living Environments (Can-ALE) dataset, and Normalized Difference Vegetation Index (NDVI, greenness) dataset) were assessed. The main outcome variables were physical activity and sedentary time as measured by a modified version of the Physical Activity for Elderly Scale. The sample consisted of adults aged 45 and older (n = 36 580, mean age 62.6 ± 10.2, 51% female). Adjusted ordinal regression models consistently demonstrated that those residing in neighbourhoods in the highest Can-ALE category (most well-connected built environment) reported more physical activity and sedentary time. For example, males aged 75+ in the highest Can-ALE category had 2 times higher odds of reporting more physical activity (OR = 2.0, 95% CI = 1.1–3.5) and 1.8 times higher odds of reporting more sedentary time (OR = 1.8, 95% CI = 1.0–3.4). Neighbourhoods with higher greenness scores were also associated with higher odds of reporting more physical activity and sedentary time. It appears that an environment characterized by higher Can-ALE and higher greenness may facilitate physical activity, but it also facilitates more leisure sedentary time in older adults; research using device measured total sedentary time, and consideration of the types of sedentary activities being performed is needed. Novelty: Middle-aged and older adults living in neighbourhoods with higher Can-ALE scores and more greenness report more physical activity and leisure sedentary time Greenness is important for physical activity and sedentary time in middle-aged adults

Digital stethoscopes in combination with telehealth allow chest sounds to be easily collected and transmitted for remote monitoring and diagnosis. Chest sounds contain important information about a newborn's cardio-respiratory health. However, low-quality recordings complicate the remote monitoring and diagnosis. In this study, a new method is proposed to objectively and automatically assess heart and lung signal quality on a 5-level scale in real-time and to assess the effect of signal quality on vital sign estimation. For the evaluation, a total of 207 10s long chest sounds were taken from 119 preterm and full-term babies. Thirty of the recordings from ten subjects were obtained with synchronous vital signs from the Neonatal Intensive Care Unit (NICU) based on electrocardiogram recordings. As reference, seven annotators independently assessed the signal quality. For automatic quality classification, 400 features were extracted from the chest sounds. After feature selection using minimum redundancy and maximum relevancy algorithm, class balancing, and hyper-parameter optimization, a variety of multi-class and ordinal classification and regression algorithms were trained. Then, heart rate and breathing rate were automatically estimated from the chest sounds using adapted pre-existing methods. The results of subject-wise leave-one-out cross-validation show that the best-performing models had a mean squared error (MSE) of 0.49 and 0.61, and balanced accuracy of 57% and 51% for heart and lung qualities, respectively. The best-performing models for real-time analysis (, 13 pages, 8 figures, 3 tables. Paper submitted and under review in IEEE Access

Model-based market segmentation analyses often involve an ordinal dependent variable as ordinal responses are frequently collected in marketing research. In the Bayesian segmentation literature, there are models for an interval- or ratio-scaled dependent variable but there is not any general model for an ordinal dependent variable. In this manuscript, the authors propose a new Bayesian procedure to simultaneously perform segmentation and ordinal regression with variable selection within each derived segment. The procedure is robust to outliers and it also provides an option to include concomitant variables that allows the simultaneous profiling of the derived segments. The authors demonstrate that the practice of treating ordinal responses as interval- or ratio-scales to apply existing Bayesian segmentation procedures can lead to very misleading results and conclusions. Through simulation studies, the authors show that the proposed procedure outperforms several benchmark Bayesian segmentation models in parameter recovery, segment retention, and segment membership prediction for such data. Finally, they provide a commercial business customer satisfaction empirical application to illustrate the usefulness of the proposed model.

This study used a survey questionnaire to investigate the state and determinants of household food security in South Africa and Botswana. In South Africa, 1557 households in two District Municipalities within Eastern Cape Province participated in the study. In Botswana, data was collected from 506 in Southern Kanye, southeast District, and Gaborone households. The paper employed Household Food Insecurity Access Scale and Prevalence to investigate the state food security across the sample population. The Household Food Insecurity Access Prevalence was used to determine the characteristics of and changes in household food insecurity of the sample households. The linear and ordinal regression analyses were carried out to outline determinants of food insecurity in the region. Findings emanating from the current study show a high prevalence of food insecurity in both countries. Meanwhile, geographical location, household size, and socio-economic infrastructure emerged as common determinants of food security for Botswana and South Africa. The paper recommends state-mediated intervention to improve access to basic socio-economic infrastructure and address unemployment rates in both countries as key areas to improve food security status.

OBJECTIVE: To determine the influence of major depressive disorders (MDD) on the cognitive function of adults living with HIV and the prevalence of neurocognitive impairment and major depressive disorders. METHODS: The study was a cross-sectional design conducted in a referral hospital in Dodoma region central Tanzania, with a sample of 397 participants attained through systematic sampling. Neurocognitive impairment was assessed by Montreal Cognitive Assessment (MoCA), while MDD was assessed using Mini International Neuropsychiatric Interview (M.I.N.I). Two-sample independent t-tests were done to compute the significance of the mean change of MoCA score between those with or without MDD. Univariate and multivariate ordinal logistic regression models were computed to determine the association between MDD and neurocognitive impairment across all cognitive domains. RESULTS: Of 397 recruited participants, 266 (66.00%) met the criteria for neurocognitive disorder and 22 (5.51%) had major depressive disorder. Participants with MDD performed significantly poorer on cognitive testing than those without MDD (mean difference = 3.74, p-value = 0.0009). Under univariate ordinal regression, impairments in the particular domains of visuospatial-executive function, attention, language, abstraction and orientation were significantly associated with MDD and abstraction (AOR: 3.922, 95% CI: 1.546–9.947, p = 0.0004) remained significant under multivariable regression. CONCLUSION: Major depressive disorder may negatively influence neurocognitive performance in both severity and pattern of presentation. Routine screening for depressive symptoms in HIV care can be beneficial for the overall patient’s outcome. Systematic and well-designed studies in the area are crucial for understanding the interaction of neuropsychiatric disorders in the HIV population.

Background and purpose Fear of falling is associated with poor physical health and influences postural stability during whole-body movement. The ability to recover from lateral balance loss is required to prevent falls; however, the relationship between lateral balance recovery and fear of falling has not been established. This study aimed to investigate whether fear of falling is associated with the stepping threshold for lateral balance recovery. Methods This study included 56 ambulatory, community-dwelling women aged 65 years or older. We determined the single-step threshold as the maximum lean magnitude normalized with body weight from which participants could be suddenly released and still recover balance using a single side step. The short-form Falls Efficacy Scale International was used as a measure of fear of falling. Results and discussion The single-step threshold significantly correlated with age (rs = -0.603) and the short-form Falls Efficacy Scale International score (rs = -0.439). Ordinal regression analysis revealed that age (odds ratio, 0.826; 95% confidence interval, 0.742-0.920) and the short-form Falls Efficacy Scale International score (odds ratio, 0.811; 95% confidence interval, 0.680-0.966) were significantly associated with the single-step threshold, such that older age and greater fear of falling each independently predicted that failure to recover balance with a single step would occur at a lower percentage of body weight. Conclusions Greater fear of falling was associated with reduced ability to recover from lateral balance loss in addition to aging. Future studies should explore whether evidence-based interventions to reduce fear of falling combined with perturbation training might lead to improved ability to recover from balance loss.

As demand increased for mental health services, especially in university counseling centers, providers have seen increasing numbers of clients. The effect of this increase on therapist caseloads is explored, with a recognition that past research on therapist caseloads lacks direct and fluctuating measures of caseload that reflect practice in naturalistic settings. Using a large dataset from a counseling center (N = 18,322), therapist caseload was conceptualized dynamically over rolling 30-day periods, using within-therapist counts of therapy sessions, unique clients seen, and the proportion of unique clients to sessions. Analysis of variance was used first to test for differences in caseload between months, years, then to test for differences between therapists (n = 173). Hierarchical linear models were constructed to examine the relationship between changes in therapist caseload across time and client outcome. Logistic and ordinal regression approaches were used to further examine this relationship for clinically significant change. Results included finding a small, but significant, effect of therapist caseload on outcome, with this finding discussed in the context of the effect sizes in the literature on therapist effects. (PsycInfo Database Record (c) 2020 APA, all rights reserved).

INTRODUCTION This study examines the health status of long-haul truck drivers (LHTDs) and more specifically smoking and obesity, in France. METHODS A total of 373 French and other LHTDs were randomly interviewed at six highway rest stops. Variables recorded were self-reported sociodemographic characteristics (age, gender, marital status, and nationality), behavioral variables (consumption of sodas, daily exercise, smoking status), weight and height. Body mass index (BMI) was also calculated. Statistical analysis was conducted using chi-squared analysis and multiple logistic ordinal regression models (MLR) were developed. RESULTS The mean age was 43.4 and the largest proportion (41.2%) of respondents were aged 40–50 years. Most respondents were French (70.3%), married (73.8%) and did not report having an active exercise routine (66.0%). Almost two-thirds of the drivers consumed daily 1–2 sodas (34.2%) or 3–4 sodas (33.2%). Based on the BMI, respondents were divided into normal (34.3%), overweight (39.8%), obese (19.5%) or morbidly obese (9.4%) categories. The mean BMI was 27.9 kg/m2 and 51.1% of LHTDs were current or active smokers. MLR analysis revealed that French LHTDs were more likely (OR=3.04; 95% CI: 1.62–5.69) to have a normal BMI compared to other drivers. Smokers were significantly more likely than non-smokers (OR=2.12; 95% CI: 1.26–3.58) to have an above normal BMI. CONCLUSIONS These results confirm that French and other LHTDs are at high risk of non-communicable diseases. They also highlight the need to address the health risks associated with smoking and obesity among LHTDs using multifaceted strategies.

Ordinal regression predicts the objects’ labels that exhibit a natural ordering, which is vital to decision-making problems such as credit scoring and clinical diagnosis. In these problems, the ability to explain how the individual features and their interactions affect the decisions is as critical as model performance. Unfortunately, the existing ordinal regression models in the machine learning community aim at improving prediction accuracy rather than explore explainability. To achieve high accuracy while explaining the relationships between the features and the predictions, we propose a new method for ordinal regression problems, namely the Explainable Ordinal Factorization Model (XOFM). XOFM uses piecewise linear functions to approximate the shape functions of individual features, and renders the pairwise features interaction effects as heat-maps. The proposed XOFM captures the nonlinearity in the main effects and ensures the interaction effects’ same flexibility. Therefore, the underlying model yields comparable performance while remaining explainable by explicitly describing the main and interaction effects. To address the potential sparsity problem caused by discretizing the whole feature scale into several sub-intervals, XOFM integrates the Factorization Machines (FMs) to factorize the model parameters. Comprehensive experiments with benchmark real-world and synthetic datasets demonstrate that the proposed XOFM leads to state-of-the-art prediction performance while preserving an easy-to-understand explainability.

The purpose of the conducted practical research presented in the paper consists in assessing the real situation from the aspect on the development of the ecotourism in the municipality of Gevgelija and confirmation of the hypothesis that ecotourism sector is a developmental concept that enables achieving sustainable economic development (Butcher,2007). A survey questionnaire consisting of 20 questions has been used. The survey was conducted of 52 respondents in the period from 1.10.2019. until 30.11.2019. on the territory of the Gevgelija. We used programs SPSS, and made Spearman test of correlation rs, and test of Ordinal Regression. The primary goal of the conducted practical research was to assess the willingness of students in the development of ecotourism. The results obtained during the practical research allow to draw conclusions and guidelines that can imply certain behaviours and activities contributing to the development of ecotourism in the Republic of Macedonia and using opportunities for achieving sustainable economic development (Hill, 2009) The basic limitations was number of respondents. The original value of this study is confirmed through the obtained results from the conducted research confirmed the fact that ecotourism as an alternative form for tourism development and a modern concept for achieving sustainable development. Benefits for the local community that are expected from the development of eco tourism include maintenance of biodiversity, benefits for the local population, a high level of responsibility of tourism enterprises and employees working in this sector, and an active role of the local population. KEY WORDS ecotourism, small and medium enterprises, nature, sustainable economic development, tourist’s needs

Background Evaluating measurement bias is vital to ensure equivalent assessment across diverse groups. One approach for evaluating test bias, differential item functioning (DIF), assesses item-level bias across specified groups by comparing item-level responses between groups that have the same overall score. Previous DIF studies of the Beck Anxiety Inventory (BAI) have only assessed bias across age, sex, and disease duration in monolingual samples. We expand this literature through DIF analysis of the BAI across age, sex, education, ethnicity, cognitive status, and test language. Methods BAI data from a sample (n = 527, mean age=61.4 ± 12.7, mean education=10.9 ± 4.3, 69.3% female, 41.9% Hispanic/Latin American) from rural communities in West Texas, USA were analyzed. Item response theory (IRT) / logistic ordinal regression DIF was conducted across dichotomized demographic grouping factors. The Mann-Whitney U test and Hedge's g standardized mean differences were calculated before and after adjusting for the impact of DIF. Results Significant DIF was demonstrated in 10/21 items. An adverse impact of DIF was not identified when demographics were assessed individually. Adverse DIF was identified for only one participant (1/527, 0.2%) when all demographics were aggregated. Limitations These results might not be generalizable to a sample with broader racial representation, more severe cognitive impairment, and higher levels of anxiety. Conclusions Minimal item-level bias was identified across demographic factors considered. These results support prior evidence that the BAI is valid for assessing anxiety across age and sex while contributing new evidence of its clinical relevance across education, ethnicity, cognitive status, and English/Spanish test language.

This article examines the interactions of interest groups and political institutions in three post-communist countries, at present EU-acceding states. Based on representative surveys of interest groups, it shows that interactions of interest groups with political institutions are more frequent than usually assumed. Nevertheless, they significantly vary across political institutions and types of groups. The direction of lobbying is somewhat different than in established democracies because of the predominance of the executive and weak legislative institutions. Group type differences demonstrate strong business dominance which might be even greater than in established democracies. Ordinal regression analyses confirm these results. However, the findings suggest that provision of political and technical information also accounts for interest groups’ access, which sheds light on the important functions that these organizations perform even in unconsolidated democracies.

In the coming three decades, world waste will be expected to increase by 70%. Coupled with the absence of waste management technologies, unregulated dumping remained a threat to sustainable development in developing countries. To this end, this study emphasized to trace socially acceptable and environmentally friendly solid waste dumpsite for the northeastern arid cities of Ethiopia. The tuneful integration of GIS and MCDA was used for the end-to-end processing of 17 variables. At the earlier, to increase alternative dumpsites and seeking for safe zone outside the cities, 20 km buffer at Semera-logia city was used as a spatial scope of the study. Fifteen key informants (KIs) from related departments of the city and Samara University were selected to validate the model result. In the end, Ordinal Regression (OR) analysis was used to test the significance of the criteria's on suitable dumpsite selection. The pair-wise comparison matrix shows that variables such as distance from surface water (11.32%), distance from the settlement (9.88%), distance from boreholes (8.87%), and aquifer productivity (8.51%) were the top weighed criteria for the study area. The weighted overlay result shows that 10.75% and 38.64% of the study area are found to be convenient and prohibited for dumping, respectively. Indeed, the KIs were agreed on 77% of the validation points and the agreement decreases from points of not-suitable to highly suitable. Except for aquifer productivity, all the 16 variables are significantly and negatively affect dumpsites suitability. This result can be further used as a preliminary database for experts to design landfill structures and also management operations for decision-makers.Implications: The identification of potential dumpsite is an important step towards maximizing efficiency and building sustainability in the arid areas. The decision-making process shall be based on the empirical investigation. The decision-making process on dumpsite selection were well thought out both the socio-economic and environmental factors. In owing process, the empirical investigation that was conducted focusing on socio-economic dimension jeopardized other components of the environments. The approach that considers both socio-economic and spatial factors was all-inclusive. Overall, decision-makers could have a good insight to work on prioritizing dumpsite alternatives and implement feasible waste management actions.

Ordinal regression methods are widely used to predict the ordered labels of data, among which support vector ordinal regression (SVOR) methods are popular because of their good generalization. In many realistic circumstances, data are collected by a distributed network. In order to protect privacy or due to some practical constraints, data cannot be transmitted to a center for processing. However, as far as we know, existing SVOR methods are all centralized. In the above situations, centralized methods are inapplicable, and distributed methods are more suitable choices. In this paper, we propose a distributed SVOR (dSVOR) algorithm. First, we formulate a constrained optimization problem for SVOR in distributed circumstances. Since there are some difficulties in solving the problem with classical methods, we used the random approximation method and the hinge loss function to transform the problem into a convex optimization problem with constraints. Then, we propose subgradient-based algorithm dSVOR to solve it. To illustrate the effectiveness, we theoretically analyze the consensus and convergence of the proposed method, and conduct experiments on both synthetic data and a real-world example. The experimental results show that the proposed dSVOR could achieve close performance to that of the corresponding centralized method, which needs all the data to be collected together.

Orange wastage through postharvest losses has contributed to food scarcity, economic loss and massive importation of food items in Nigeria. The research was mainly carried out to investigate the determinants of orange postharvest losses among orange farmers in Konshisha Local Government of Benue State, North Central geopolitical zone of Nigeria. Primary data was collected from the orange farmers using structured questionnaires and key informant interviews. Descriptive statistics and Ordinal Regression model were used to analyse the data collected. The quantity lost was perceived at six (6) categories. The results revealed that most (63.7%) of the farmers were above 34 years of age. Also the majority (95.1%) were male, while 55.3% of the respondents’ farm size was relatively large with 200 and above stands of orange. The farmers’ literacy level was 73.6%. Those that belonged to farmers groups were 39.5. Further results established the use of probit link function in the ordinal regression modelling and that the significant factors affecting orange postharvest losses in the area are farmer’s lack of education and farmers not belonging to any association or group. The only significant covariate with the postharvest loss quantity of orange is farm size. The test of parallel lines established that, the location parameters (slope coefficients) are the same across response categories.

Housing across the world has remained an indispensable phenomenon that affects every facet of humans. Its relevance is so evident that it imparts on man’s socio-physical and mental welfare irrespective of his socio-economic status, colour or creed. The correct socio-economic position is linked to people seeking inexpensive and decent housing. This study assessed the socio-economic characteristics of households and their housing condition in the Bauchi metropolis. The study adopted a quantitative approach where 380 questionnaires were administered to house hold-heads in this study. The households were sampled through stratified random sampling to generate data on their socio-economic characteristics, housing conditions, existing facilities and amenities, physical and environmental characteristics. The data collected were subjected to descriptive statistics with mean ranking and ordinal regression to examine the significance of the various variables. The findings of the study revealed that socio-economic characteristics affect housing conditions in the study area. The study also revealed that compared with the medium and high-density areas where the housing situation and all basic infrastructures are fair and foul, the low-density area had its housing condition with all basic infrastructures in good condition. It was recommended that the government provide adequate social facilities as a matter of urgency, renovate the declining ones, and implement development control standards in the medium and high-density areas. Furthermore, individuals should incorporate a good maintenance culture for their property to improve housing and environmental characteristics.

The so-called proportional odds assumption is popular in cumulative, ordinal regression. In practice, however, such an assumption is sometimes too restrictive. For instance, when modeling the perception of boar taint on an individual level, it turns out that, at least for some subjects, the effects of predictors (androstenone and skatole) vary between response categories. For more flexible modeling, we consider the use of a ‘smooth-effects-on-response penalty’ (SERP) as a connecting link between proportional and fully non-proportional odds models, assuming that parameters of the latter vary smoothly over response categories. The usefulness of SERP is further demonstrated through a simulation study. Besides flexible and accurate modeling, SERP also enables fitting of parameters in cases where the pure, unpenalized non-proportional odds model fails to converge.

The purpose of this paper is to introduce a new distance metric learning algorithm for ordinal regression. Ordinal regression addresses the problem of predicting classes for which there is a natural ordering, but the real distances between classes are unknown. Since ordinal regression walks a fine line between standard regression and classification, it is a common pitfall to either apply a regression-like numerical treatment of variables or underrate the ordinal information applying nominal classification techniques. On a different note, distance metric learning is a discipline that has proven to be very useful when improving distance-based algorithms such as the nearest neighbors classifier. In addition, an appropriate distance can enhance the explainability of this model. In our study we propose an ordinal approach to learning a distance, called chain maximizing ordinal metric learning. It is based on the maximization of ordered sequences in local neighborhoods of the data. This approach takes into account all the ordinal information in the data without making use of any of the two extremes of classification or regression, and it is able to adapt to data for which the class separations are not clear. We also show how to extend the algorithm to learn in a non-linear setup using kernel functions. We have tested our algorithm on several ordinal regression problems, showing a high performance under the usual evaluation metrics in this domain. Results are verified through Bayesian non-parametric testing. Finally, we explore the capabilities of our algorithm in terms of explainability using the case-based reasoning approach. We show these capabilities empirically on two different datasets, experiencing significant improvements over the case-based reasoning with the traditional Euclidean nearest neighbors.

Ordinal regression is used to model the ordinal response variable as functions of several explanatory variables. The most commonly used model for ordinal regression is the proportional odds model (POM). The classical technique for estimating the unknown parameters of this model is the maximum likelihood (ML) estimator. However, this method is not suitable for solving problems with extreme observations. A robust regression method is needed to handle the problem of extreme points in the data. This study proposes Huber M-estimator as a robust method to estimate the parameters of the POM with a logistic link function and polytomous explanatory variables. This study assesses ML estimator performance and the robust method proposed through an extensive Monte Carlo simulation study conducted using statistical software, R. Measurement for comparisons are bias, RMSE, and Lipsitzs' goodness of fit test. Various sample sizes, percentages of contamination, and residual standard deviations are considered in the simulation study. Preliminary results show that Huber estimates provide the best results for parameter estimation and overall model fitting. Huber's estimator has reached a 50% breakdown point for data containing extreme points that are quite far from most points. In addition, the presence of extreme points that have only a distance of two times far from most points has no major impact on ML estimates. This means that the estimates for ML and Huber may yield the same results if the model's residual values are between -2 and 2. This situation may also occur for data with a percentage of contamination below 5%.

Background and aim Frailty Phenotype (FP) is very common in Heart Failure (HF) and both syndromes worsen one another. The aim of this study is to first describe FP in a sample of Portuguese HF patients, and to analyse its association with nutritional and clinical statuses, namely muscle mass, obesity and functional class. Methods and results In this cross-sectional study, a sample of 136 heart failure outpatients (24-81 years, 33.8% women) was randomly selected from the appointments’ listings of a HF and Transplant clinic in a Portuguese University Hospital. FP was assessed according to Fried et al.; muscle mass was estimated from mid-upper arm muscle circumference; weight status was assessed using body mass index; HF functional classes were registered. The association between participants’ characteristics and FP categories was analysed using logistic ordinal regression. The frequency of pre-frailty and frailty is 57.4% and 15.4%, respectively. Within frail individuals, 52.4% were under the age of 65. In multivariable analysis, frailty was positively associated with being 70 or older (OR = 3.44) and obese (OR = 2.66), and negatively associated with muscle mass (OR = 0.77) and with being at class I (OR = 0.14) or class II (OR = 0.29) of HF functional classes. Conclusion Muscle mass seems to be an important predictor of frailty in HF patients and should be taken into account when designing intervention plans that allow for reverting or modifying frailty and pre-frailty. Younger patients should be monitored for the presence and evolution of FP.

Background and Purpose: Understanding factors which affect the level of family well-being is important as it contributes to effective decision making among the policymakers to improve the family lives as well as to strengthen the family institution. Accordingly, this line of research is gaining attention. This study develops an ordinal regression model which identifies demographic, economic and social factors that are significant in explaining the status of family well-being. Methodology: Data involving 2,808 respondents from a nationwide survey conducted by the National Population and Family Development Board of Malaysia in 2011 were used in this study. Ordinal regression model was implemented to describe the three levels of family well-being. Findings: The national survey reported that high level of family well-being was experienced by 76.3 per cent of the respondents, followed by moderate (18.4%) and low (5.3%). The fitted ordinal regression model found that ethnic background, family relationship, community relationship, health and safety levels, economic situation of the family, religious practice, housing, and environment are significantly related to family well-being. Meanwhile, it was found that the level of income is not a significant factor in determining the level of family well-being. Contributions: There are a limited number of studies on the application of ordinal regression for modelling the level of family well-being, particularly with covariates involving the demographic and social characteristics of the respondents. This study fills in the gap in the literature where the ordinal regression model provides useful information for policymakers to enhance the status of family well-being in Malaysia via various policy initiatives. Keywords: Family well-being, Ordinal Regression Model, ordinal data, Proportional Odds Model. Cite as: Muhammad Sapri, N. A., Ibrahim, K., Abu Bakar, M. A., & Mohd Ariff, N. (2021). Ordinal regression for modelling the family well-being among the Malaysians. Journal of Nusantara Studies, 6(2), 424-447. http://dx.doi.org/10.24200/jonus.vol6iss2pp424-447

Objectives: To evaluate the demographic and clinical characteristics of patients with Behcet's syndrome (BS), and to define their associations with the presence of major organ involvement (MOI).Method: Medical records of 2118 patients (964 males, 1154 females) were analysed retrospectively. MOI was defined as the presence of at least one of vascular, eye, nervous, or gastrointestinal system involvement. Univariable and multivariable binary and ordinal logistic regression analyses were applied to assess the factors that were potentially associated with MOI.Results: The mean ± sd age at diagnosis was 30.5 ± 9.4 years. Genital ulcer and joint involvement were more common in females (both p < 0.001), while MOI was more frequent in males (p < 0.001). Genital ulcer (p < 0.001) and vascular involvement (p = 0.006) were more common in patients with a younger age at diagnosis, while joint involvement was more common in older patients. A total of 1097 patients (51.8%) had at least one MOI, 322 (15.2%) at least two MOIs, and 48 (2.3%) at least three MOIs. Male gender, smoking history, and absence of genital ulcer were significantly associated with MOI in multivariable binary logistic regression. Multivariable ordinal regression analyses confirmed the association between MOI and male gender and smoking, but not the protective effect of genital ulcers. In both regression analyses, we found no significant effects of age, human leucocyte antigen-B51, skin involvement, or joint involvement on MOI.Conclusion: Male gender and positive smoking history have a significant influence on the presence of MOI in patients with BS.

The assessment of pedestrian facilities largely depends on the Level of Service (LOS) calculation taking into account the effective width and the pedestrians’ flows. This research focuses on the investigation of pedestrians’ sociodemographic attributes (gender, age), qualitative components (perceived comfort) and the pedestrian streets’ characteristics for LOS estimation as perceived by pedestrians. Α questionnaire-based survey (including 200 interviewees) was carried out at the central pedestrianized network in the city of Ioannina, Greece. An ordinal regression model was developed based on the questionnaire’ data. According to the findings females and elderly pedestrians tend to perceive lower LOS than males and younger pedestrians respectively. Furthermore, the level of perceived comfort in the impact area of the urban central network is an important determinant for the estimation of perceived LOS. These findings can be effectively utilized in the design of sustainable mobility policy recommendations and the promotion of active transport in an urban central area.

Ordinal variables, while extremely common in Psychology, are almost exclusively analysed with statistical models that falsely assume them to be metric. This practice can lead to distorted effect size estimates, inflated error rates, and other problems. We argue for the application of ordinal models that make appropriate assumptions about the variables under study. In this tutorial article, we first explain the three major ordinal model classes; the cumulative, sequential and adjacent category models. We then show how to fit ordinal models in a fully Bayesian framework with the R package brms, using data sets on stem cell opinions and marriage time courses. Two appendices provide detailed mathematical derivations of the models, and a third practical example that connects ordinal models to Signal Detection Theory with confidence rating data. Ordinal models provide better theoretical interpretation and numerical inference from ordinal data, and we recommend their widespread adoption in Psychology.

Public stigmatisation has been identified as a barrier to seeking therapy amongst people with paedophilia (PWP). This study piloted a public stigmatism intervention towards non-offending paedophilia and compared its effectiveness with a public stigmatism intervention towards alcoholism. The overall aim was to reduce self-reported stigmatism towards PWP and assess the comparative effectiveness of the stigmatism intervention towards PWP with the alcoholism stigmatism intervention. A 2x2 mixed effects design was utilised to allow for the comparison of the stigmatism interventions. 80 participants were recruited opportunistically online and were provided with a text-based educational intervention combined with a counter-stereotypical exemplar video. This data was analysed using Bayesian Cumulative Probit Ordinal Regression Models. There was an effect of condition which indicated that stigma inventory scores towards PWP were consistently higher pre-intervention, estimate = 0.44, posterior probability = .99. No other estimates had 95% credible intervals (C.I) which did not include zero, making the results inconclusive. Despitethis, the hypothesis distributions indicated that stigma inventory estimates tended towards the hypothesised results. The Perceived Social Distance model indicated that there may have been an effect of the intervention that was unique to the PWP condition although the C.I’s for this hypothesis, and for all others, also included zero. The Social Distance model estimates did not produce any of the hypothesised results and none of the C.I’s included zero. Overall, this study found that a public stigmatism intervention has the potential to change stigmatising beliefs towards PWP and influence the perceived opinions of peers towards PWP. It also contests the proposed differences in PWP and alcoholism social distancing. Notably it provides a tentative first step in the development of a public stigmatism intervention which could potentially increase PWP participation in formal therapeutic services.

This study introduces a coefficient of discrimination for use with nominal and ordinal regression models. Computation of the coefficient is demonstrated with data from the Pew Research Center’s 25th Anniversary of the Web Omnibus Survey pertaining to cell/home phone ownership, where the coefficient of discrimination indicates that respondent age and gender increased the probability of a correct versus incorrect classification by 13.9%. Additionally, the coefficient is compared to existing coefficients.

Generally, modeling poverty aims to obtain the best criteria for assessing poverty status. There are two approaches to model the factors that affect poverty, namely consumption approach and discrete choice model. The advantage of the discrete choice model compared to the consumption approach is that the discrete choice model provides a probabilistic estimate for classifying samples into different poverty categories. This study aims to examined how the factors that affect poverty in Yogyakarta through Regularized Ordinal Regression with elastic net approach both for parallel, non-parallel, and semi-parallel models. The data used in this study is Susenas March 2018 for Yogyakarta provinces. The result of this study shows that the best discrete choice model for Yogyakarta’s modelling is the parallel model. Households that live in villages, have a large number of household members, are headed by women, have elderly household heads, have low education, and work in the primary sector tend to be more vulnerable to poverty. Therefore, a simultaneous policy with inclusive economic development is needed to reduce cross-border, cross-gender, and cross-sector inequality

Teachers´ and special educators’ attitudes towards research-based work differ but remains poorly understudied. This study predicts Swedish teachers´ and special educators´ attitudes towards (a) research-based work and (b) inviting researchers to schools. The data comes from a survey with a convenience sample of more than 300 teachers and special educators. The analysis combines methods such as: factor analysis, bootstrapped linear regression, and ordinal regression. Theoretically, this study derives predictions from professional capital theory. First, the results show how teachers and special educators differ in attitudes towards research-based work. Second, the results show how teachers with greater social capital have a higher probability of willingness to invite researchers to the school. This study concludes support for professional capital theory.

Driver mental fatigue leads to thousands of traffic accidents. The increasing quality and availability of low-cost electroencephalogram (EEG) systems offer possibilities for practical fatigue monitoring. However, non-data-driven methods, designed for practical, complex situations, usually rely on handcrafted data statistics of EEG signals. To reduce human involvement, we introduce a data-driven methodology for online mental fatigue detection: self-weight ordinal regression (SWORE). Reaction time (RT), referring to the length of time people take to react to an emergency, is widely considered an objective behavioral measure for mental fatigue state. Since regression methods are sensitive to extreme RTs, we propose an indirect RT estimation based on preferences to explore the relationship between EEG and RT, which generalizes to any scenario when an objective fatigue indicator is available. In particular, SWORE evaluates the noisy EEG signals from multiple channels in terms of two states: shaking state and steady state. Modeling the shaking state can discriminate the reliable channels from the uninformative ones, while modeling the steady state can suppress the task-nonrelevant fluctuation within each channel. In addition, an online generalized Bayesian moment matching (online GBMM) algorithm is proposed to online-calibrate SWORE efficiently per participant. Experimental results with 40 participants show that SWORE can maximally achieve consistent with RT, demonstrating the feasibility and adaptability of our proposed framework in practical mental fatigue estimation.

Project risk management is crucial for project success and for achieving short-term and long-term project goals. This study examines the linkage between the management of project risks and project flexibility for information technology projects in Financial Services. A conceptual framework establishing the link between project risks, project flexibility, project performance, and business success, with project innovativeness as a moderating variable, has been introduced. To test the model, data were collated from over 400 managers working in Financial Services projects. The empirical outcomes through a Ordinal regression analysis demonstrate a substantial association between the management of project risks, project flexibility, and success of projects. Project innovativeness moderates the effects of project risks and project flexibility on project performance. Furthermore, managing project risks is vital to reduce the likelihood of failures in projects. This paper enriches existing research by applying a contingency perspective to project risk management and provides practical guidance for managing risks in projects professionally and also the relevancy of project flexibility.

Purpose: To investigate the relationship between the training load (TL = rate of perceived exertion × training time) and wellness index (WI) in soccer. Methods: The WI and TL data were recorded from 28 subelite players (age = 20.9 [2.4] y; height = 181.0 [5.8] cm; body mass = 72.0 [4.4] kg) throughout the 2017/2018 season. Predictive models were constructed using a supervised machine learning method that predicts the WI according to the planned TL. The validity of our predictive model was assessed by comparing the classification’s accuracy with the one computed from a baseline that randomly assigns a class to an example by respecting the distribution of classes (B1). Results: A higher TL was reported after the games and during match day (MD)-5 and MD-4, while a higher WI was recorded on the following days (MD-6, MD-4, and MD-3, respectively). A significant correlation was reported between daily TL (TLMDi) and WI measured the day after (WIMDi+1) (r = .72, P

PURPOSE To report changes in denture satisfaction and oral health-related quality of life (OHRQoL) in edentulous patients treated with two-implant mandibular overdentures (IODs) over a follow-up of 8 to 10 years. MATERIALS AND METHODS This is a follow-up of a previous study carried out between 1997 and 2005. The patients were originally randomly divided into one group receiving IODs and another group who had their conventional mandibular dentures relined (RCD group). The RCD group was offered and received IODs at the 2-year follow-up. The participants completed a self-administered questionnaire containing demographics, 15 variables related to denture satisfaction, and 20 questions from the Oral Health Impact Profile (OHIP-20). Changes over time were analyzed using multilevel linear models for denture satisfaction and multilevel ordinal regression analyses for OHIP-20 variables. Comparisons between groups were analyzed using Mann-Whitney U test for ordinal and t test for metric data. RESULTS Disregarding patients who passed away during follow-up, the 29 responders represented a response rate of 76%. The degree of denture satisfaction and the OHIP-20 scores remained high and stable in the IOD group over the 10-year observation period for all but one variable. The same factors showed only a modest improvement in the RCD group for the first 2 years; however, during the subsequent 8 years of the observation period (after receiving IODs), denture satisfaction and OHIP-20 scores improved to the same level as the original IOD group. CONCLUSION The positive effect on denture satisfaction and OHRQoL of edentulous patients treated with two-implant mandibular overdentures remained unchanged 8 to 10 years after treatment.

Social media are deemed influential in making decisions and seeking advice. Due to their explosive growth as critical channels for information, their content can trigger a place visit, a change of transport mode or destination, or plans’ cancellation. The main objective of this paper is to investigate the influence of social media on users’ activity and mobility planning. Responses of 738 participants in a digital survey were used to formulate ordinal regression models. The developed models determine the contribution of users’ demographic characteristics, travel characteristics and social media usage to mobility decisions after using social media as a source of information. These decisions were expressed in two dependent variables; (i) the impact of social media use in activity and mobility planning; (ii) the impact of the proposed transport mode by social media information, on mode choice. Analysis of the results indicated that the models, which considered all the characteristics together, could better predict the two variables.

Purpose: The purpose of this study was to determine whether positive changes in self-reported attitudes toward disability were retained after students had participated in a 12-week voluntary community-based exercise programme for young people with disability. Method: Sixty-eight physiotherapy and exercise physiology students (45 women, 23 men; mean age 21 y) volunteered. Each student was matched with a young person with disability, and the pair exercised together twice a week for 12 weeks at their local gymnasium. The primary outcome measure was self-reported attitude toward disability. Secondary outcomes were self-reported professional skills competency, confidence, future work intentions, and motivations for volunteering. The outcomes were measured at Weeks 0, 13, and 24. The data were analyzed using generalized linear mixed-effects models, ordinal regression, and linear regression. Results: The students’ attitudes toward disability were more positive at Weeks 13 and 24 than at baseline, but positive changes were not fully retained by Week 24. Small statistically significant reductions occurred after Week 13. Self-reported skills competencies, confidence, and future intentions of working in disability were higher at Weeks 13 and 24. Students’ motivation for volunteering was wanting to support others, wanting to learn, and perceived positive career effects. Conclusions: Positive changes in students’ attitudes toward disability were generally retained 3 months after volunteering in a community-based exercise programme.

Objective The prevalence of social media for sharing personal thoughts makes it a viable platform for the assessment of suicide risk. However, deep learning models are not able to capture the diverse nature of linguistic choices and temporal patterns that can be exhibited by a suicidal user on social media and end up overfitting on specific cues that are not generally applicable. We propose Adversarial Suicide assessment Hierarchical Attention (ASHA), a hierarchical attention model that employs adversarial learning for improving the generalization ability of the model. Material and Methods We assess the suicide risk of a social media user across 5 levels of increasing severity of risk. ASHA leverages a transformer-based architecture to learn the semantic nature of social media posts and a temporal attention-based long short-term memory architecture for the sequential modeling of a user’s historical posts. We dynamically generate adversarial examples by adding perturbations to actual examples that can simulate the stochasticity in historical posts, thereby making the model robust. Results Through extensive experiments, we establish the face-value of ASHA and show that it significantly outperforms existing baselines, with the F1 score of 64%. This is a 2% and a 4% increase over the ContextBERT and ContextCNN baselines, respectively. Finally, we discuss the practical applicability and ethical aspects of our work pertaining to ASHA, as a human-in-the-loop framework. Discussion and Conclusions Adversarial samples can be helpful in capturing the diverse nature of suicidal ideation. Through ASHA, we hope to form a component in a larger human-in-the-loop infrastructure for suicide risk assessment on social media.

Many real-world datasets are labeled with natural orders, i.e., ordinal labels. Ordinal regression is a method to predict ordinal labels that finds a wide range of applications in data-rich domains, such as natural, health and social sciences. Most existing ordinal regression approaches work well for independent and identically distributed (IID) instances via formulating a single ordinal regression task. However, for heterogeneous non-IID instances with well-defined local geometric structures, e.g., subpopulation groups, multi-task learning (MTL) provides a promising framework to encode task (subgroup) relatedness, bridge data from all tasks, and simultaneously learn multiple related tasks in efforts to improve generalization performance. Even though MTL methods have been extensively studied, there is barely existing work investigating MTL for heterogeneous data with ordinal labels. We tackle this important problem via sparse and deep multi-task approaches. Specifically, we develop a regularized multi-task ordinal regression (MTOR) model for smaller datasets and a deep neural networks based MTOR model for large-scale datasets. We evaluate the performance using three real-world healthcare datasets with applications to multi-stage disease progression diagnosis. Our experiments indicate that the proposed MTOR models markedly improve the prediction performance comparing with single-task ordinal regression models., Comment: 21 pages, 3 figures

Objective To describe patterns of use of self-administered nitrous oxide (N2O) during labor and to determine if maternal and neonatal process and outcome measures differ for women who use N2O compared to women who do not use N2O. Design Retrospective, full-census, observational cohort. Setting An upper midwestern U.S., urban, 75-bed quaternary perinatal center with more than 5,000 annual births. Participants The participants included two groups of women: 400 who used N2O during labor and a comparison group of 6,733 who met N2O eligibility but did not use N2O. Methods We used descriptive statistics to examine patterns of use of N2O during labor between January 2015 and March 2017. We examined associations of N2O with process (length of first and second stages of labor, time from hospital admission to birth, time from birth to hospital discharge, and total length of stay) and outcome measures (shoulder dystocia, instrumentation, vaginal lacerations, Apgar scores at 5 minutes, nursery disposition) using multivariate linear, logistic, and ordinal regression models. Results Three percent (12/400) of women who used N2O discontinued because of side effects. Among participants with vaginal births who used N2O, 17.6% (62/352) used N2O as the only form of pain medication during labor. We found no significant differences in maternal and neonatal outcome measures between the two groups. Among the process measures examined, we found a mean 2-hour-longer time from admission to birth and total length of stay in the N2O group (p Conclusion Most participants who used N2O (290/352, 82.3%) transitioned to other pain modalities during labor. Maternal and neonatal process and outcome measures were comparable relative to other pain management modalities, with the exception of longer time durations for two measures.

Demonetization is the process of declining the use of currency from circulation by the government or monetary authorities in a country. This research paper analyses the efficacy of Indian demonetization from common public perspectives, the policy that the government of India has implemented to fight against black money, drying the financial roots of terrorism, and direct the civilization towards digital transactions and a cashless economy. A field survey was conducted in Karnataka and Kerala’s coastal region by distributing a structured questionnaire among the common public to generate the data. The authors run descriptive statistics and ordinal regression analysis to obtain the result for the study’s objectives. The descriptive statistics result found that demonetization increased the number of bank account holders in India. There is not much impact of demonetization on controlling evasion of tax and illegal investments of black money, and the policy adversely affects regular business in the country. The findings from ordinal regression reveal that the time frame was given to the public to demonetize their old notes were sufficient; money circulation was well planned at the time of demonetization. The policy implemented at the right time and the common public, despite facing enormous challenges while purchasing goods and services at the time of demonetization, considers that demonetization implementation was effective. JEL Classification Codes: E5, E6, E7, E58, E60.

The authors describe a methodology of automated making soil erosion class maps using ordinal logistic regression, where relief characteristics are predictors. For mapping, a probabilistic approach is used, in which it is proposed to evaluate the possibility of belonging to several arbitrary erosion classes for each point in space

Yilin Ning,1,2 Peh Joo Ho,3,4 Nathalie C Støer,5,6 Ka Keat Lim,7,8 Hwee-Lin Wee,3,9 Mikael Hartman,1– 3,10 Marie Reilly,11 Chuen Seng Tan3 1NUS Graduate School for Integrative Sciences and Engineering, National University of Singapore, Singapore, Singapore; 2Yong Loo Lin School of Medicine, Department of Surgery, National University of Singapore and National University Health System, Singapore, Singapore; 3Saw Swee Hock School of Public Health, National University of Singapore and National University Health System, Singapore, Singapore; 4Genome Institute of Singapore, Singapore, Singapore; 5Norwegian National Advisory Unit on Women’s Health, Oslo University Hospital, Oslo, Norway; 6Department of Research, Cancer Registry of Norway, Oslo, Norway; 7Department of Population Health Sciences, School of Population Health & Environmental Sciences (SPHES), Faculty of Life Sciences & Medicine, King’s College London, London, UK; 8Programme in Health Services & Systems Research, Duke-NUS Medical School, Singapore, Singapore; 9Department of Pharmacy, Faculty of Science, National University of Singapore, Singapore, Singapore; 10Department of Surgery, National University Hospital, Singapore, Singapore; 11Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, SwedenCorrespondence: Chuen Seng TanSaw Swee Hock School of Public Health, National University of Singapore and National University Health System, Tahir Foundation Building, 12 Science Drive 2, #10-01, Singapore, 117549, SingaporeTel +65-66013206Email ephtcs@nus.edu.sgPurpose: Assessing the clinical importance of an exposure effect on a quality of life (QoL) score often requires quantifying the effect in terms of a difference in scores. Using the linear regression model (LRM) for this purpose assumes the ordinal score is a proxy for an underlying continuous variable, but the analysis offers no assessment for the validity of the assumption. We propose an approach that assesses the proxy assumption and estimates the exposure effect by using the cumulative link model (CLM).Patients and methods: CLM is a well-established regression model that assumes an ordinal score is an ordered category generated from applying thresholds to a latent continuous variable. Our approach assesses the proxy assumption by testing whether these thresholds are equidistant. We compared the performance of CLM and LRM using simulated ordinal data and illustrated their application to the effect of time since diagnosis on five subscales of fatigue among breast cancer survivors measured using the Multidimensional Fatigue Inventory.Results: CLM had good performance in estimating the difference in means with simulated ordinal data satisfying the proxy assumption, even when the outcome had only a few categories. When the proxy assumption was inadequate, both the CLM and LRM had biased estimates with poor coverage. The proxy assumption was appropriate for four ofthe five subscales in our real data application to fatigue scores, which highlighted the importance of assessing the proxy assumption to avoid reporting invalid estimates in terms of the difference in scores.Conclusion: The proxy assumption is critical to the interpretation of the exposure effect on the difference in mean QoL scores. CLM offers a valid test for the presence of an association, a method for assessing the proxy assumption, and when the assumption is adequate, an assessment for clinical significance using the difference in means.Keywords: cumulative link model, ordered probit model, ordinal outcome, ordinal regression, probit link, quality of life

OBJECTIVE: We developed a diagnostic overuse index that identifies hospitals with high levels of diagnostic intensity by comparing negative diagnostic testing rates for common diagnoses. METHODS: We prospectively identified candidate overuse metrics, each defined by the percentage of patients with a particular diagnosis who underwent a potentially unnecessary diagnostic test. We used data from seven states participating in the State Inpatient Databases. Candidate metrics were tested for temporal stability and internal consistency. Using mixed-effects ordinal regression and adjusting for regional and hospital characteristics, we compared results of our index with three Dartmouth health service area-level utilization metrics and three Medicare county-level cost metrics. RESULTS: The index was comprised of five metrics with good temporal stability and internal consistency. It correlated with five of the six prespecified overuse measures. Among the Dartmouth metrics, our index correlated most closely with physician reimbursement, with an odds ratio of 2.02 (95% CI, 1.11-3.66) of being in a higher tertile of the overuse index when comparing tertiles 3 and 1 of this Dartmouth metric. Among the Medicare county-level metrics, our index correlated most closely with standardized costs of procedures per capita, with an odds ratio of 2.03 (95% CI, 1.21-3.39) of being in a higher overuse index tertile when comparing tertiles 3 and 1 of this metric. CONCLUSIONS: We developed a novel overuse index that is preliminary in nature. This index is derived from readily available administrative data and shows some promise for measuring overuse of diagnostic testing at the hospital level.

With the development of social media sites, user credit grading, which served as an important and fashionable problem, has attracted substantial attention from a slew of developers and operators of mobile applications. In particular, multi-grades of user credit aimed to achieve (1) anomaly detection and risk early warning and (2) personalized information and service recommendation for privileged users. The above two goals still remained as up-to-date challenges. To these ends, in this article, we propose a novel regression-based method. Technically speaking, we define three natural ordered categories including BlockList , GeneralList , and AllowList according to users’ registration and behavior information, which preserve both the global hierarchical relationship of user credit and the local coincident features of users, and hence formulate user credit grading as the ordinal regression problem. Our method is inspired by KDLOR ( kernel discriminant learning for ordinal regression ), which is an effective and efficient model to solve ordinal regression by mapping high-dimension samples to the discriminant region with supervised conditions. However, the performance of KDLOR is fragile to the extreme imbalanced distribution of users. To address this problem, we propose a robust sampling model to balance distribution and avoid overfit or underfit learning, which induces the triplet metric constraint to obtain hard negative samples that well represent the latent ordered class information. A step further, another salient problem lies in ambiguous samples that are noises or located in the classification boundary to impede optimized mapping and embedding. To this problem, we improve sampling by identifying and evading noises in triplets to obtain hard negative samples to enhance robustness and effectiveness for ordinal regression. We organized training and testing datasets for user credit grading by selecting limited items from real-life huge tables of users in the mobile application, which are used in similar problems; moreover, we theoretically and empirically demonstrate the advantages of the proposed model over established datasets.

Osteoarthritis (OA) is a common form of knee arthritis which causes significant disability and is threatening to plague patient’s quality of life. Although this chronic condition does not lead to fatality, still there exists no known cure for OA. Diagnosis of OA can be confirmed primarily based on radiographic findings. Being a progressive disease, early identification of OA is crucial for clinical interventions to curtail the OA degeneration. Kellgren-Lawrence (KL) grading system has been traditionally employed to assess the knee OA severity. Due to the recent advancements of deep learning in computer vision, more studies have employed deep neural network in automatically predicting KL grade from plain knee joint radiograph. However, these studies treat KL grading as a multi-class classification task and ignore the inherent ordinal nature within the KL grades. In this study, we propose an ordinal regression module for neural networks to treat KL grading as an ordinal regression task. Our module takes an input from neural network and produces 4 cut-points to partition the prediction space into 5 respective KL grades. The proposed model is optimized by a cumulative-link loss function. Performance of the model is evaluated against various notable neural networks and significant improvements on the knee OA KL grade prediction were demonstrated.

Purpose This paper aims to investigate how the students perceive the cleaning of the university, from an in-house and outsourced perspective. Design/methodology/approach This research is based on a survey conducted in the different university buildings with a total of 240 students. The survey was then analysed through an ordinal regression. Findings The ordinal regression indicates a statistically significant result were student are more satisfied with the cleaning performed by the outsourced service provider. Research limitations/implications The research in this paper is limited to one Swedish university. But, the overall strategies for how to organise the cleaning service at the university do address all universities. However, the research is limited and more research has to be performed. Practical implications The insight the paper provides regarding how the students perceive the cleaning service at the university gives direct implications to university in relation to how to consider the cleaning service as an important factor. Originality/value It provides the first study from a student perspective on the question of cleaning of the university, when previous studies have indicated cleaning as an important function not least to student’s performance and academic results.