Multiple Performance Measures Research Articles

Comparison of vertical jump and sprint performances between 3 × 3 and 5 × 5 elite professional male basketball players.

Given its fast-growing popularity and unique on-court competitive demands, 3 × 3 basketball has captured a considerable amount of attention over recent years. However, unlike research focused on studying 5 × 5 basketball players, there is a lack of scientific literature focused on examining countermovement vertical jump (CMJ) and sprint performance characteristics of 3 × 3 athletes. Thus, the purpose of the present study was to compare force-time metrics during both eccentric and concentric phases of the CMJ and acceleration and deceleration capabilities between 3 × 3 and 5 × 5 top-tier professional male basketball athletes. Ten 3 × 3 and eleven 5 × 5 professional basketball players volunteered to participate in the present study. Upon completion of a standardized warm-up, each athlete performed three maximum-effort CMJs, followed by two 10 m sprints. A uni-axial force plate system sampling at 1,000 Hz was used to analyze CMJ force-time metrics and a radar gun sampling at 47 Hz was used to derive sprint acceleration-deceleration measures. Independent t-tests and Hedge's g were used to examine between-group statistically significant differences (p < 0.05) and effect size magnitudes. The findings of the present study reveal that 3 × 3 and 5 × 5 professional male basketball players tend to display similar neuromuscular performance characteristics as no significant differences were observed in any force-time metric during both eccentric and concentric phases of the CMJ (g = 0.061-0.468). Yet, prominent differences were found in multiple measures of sprint performance, with large effect size magnitudes (g = 1.221-1.881). Specifically, 5 × 5 basketball players displayed greater average and maximal deceleration and faster time-to-stop than their 3 × 3 counterparts. Overall, these findings provide reference values that sports practitioners can use when assessing athletes' CMJ and sprint performance capabilities as well as when developing sport-specific training regimens to mimic on-court competitive demands.

Enhancing the performance of hybrid wave-wind energy systems through a fast and adaptive chaotic multi-objective swarm optimisation method

Hybrid offshore renewable energy platforms have been proposed to optimise power production and reduce the levelised cost of energy by integrating or co-locating several renewable technologies. One example is a hybrid wave-wind energy system that combines offshore wind turbines with wave energy converters (WECs) on a single floating foundation. The design of such systems involves multiple parameters and performance measures, making it a complex, multi-modal, and expensive optimisation problem. This paper proposes a novel, robust and effective multi-objective swarm optimisation method (DMOGWA) to provide a design solution that best compromises between maximising WEC power output and minimising the effect on wind turbine nacelle acceleration. The proposed method uses a chaotic adaptive search strategy with a dynamic archive of non-dominated solutions based on diversity to speed up the convergence rate and enhance the Pareto front quality. Furthermore, a modified exploitation technique (Discretisation Strategy) is proposed to handle the large damping and spring coefficient of the Power Take-off (PTO) search space. To evaluate the efficiency of the proposed method, we compare the DMOGWA with four well-known multi-objective swarm intelligence methods (MOPSO, MALO, MODA, and MOGWA) and four popular evolutionary multi-objective algorithms (NSGA-II, MOEA/D, SPEA-II, and PESA-II) based on four potential deployment sites on the South Coast of Australia. The optimisation results demonstrate the dominance of the DMOGWA compared with the other eight methods in terms of convergence speed and quality of solutions proposed. Furthermore, adjusting the hybrid wave-wind model’s parameters (WEC design and PTO parameters) using the proposed method (DMOGWA) leads to a considerably improved power output (average proximate boost of 138.5%) and a notable decline in wind turbine nacelle acceleration (41%) throughout the entire operational spectrum compared with the other methods. This improvement could lead to millions of dollars in additional income per year over the lifespan of hybrid offshore renewable energy platforms.

Physical Education Teaching Quality Evaluation Method Using Mobile Edge Computing in the Online and Offline Environment

The development of different technology has impacted the different stages of human life. Additionally, with the implication of computing technology sports and physical education can be improved. The use of mobile edge technology aids in gathering precise data related to physical education. Therefore, a specific improvement is possible. This following analysis has looked into the factor of mobile edge computing that aids in the evolution of physical education teaching. Moreover, the study has focused on developing an appropriate path for the integration of MEC into PE education. By utilising Mobile Edge Computing (MEC) technologies in both online and off-line learning contexts, this study offers a thorough way for assessing the quality of sports teaching. The suggested approach integrates multiple characteristics and performance measures to evaluate the efficacy of physical education instruction and participation, taking into consideration the changing educational landscape.

Thirty years of longitudinal talent development research: a systematic review and meta-aggregation

ABSTRACT Talent pathways are longitudinal and multidimensional in nature offering developmental environments for athletes that incorporate multiple processes at multiple timepoints. Recent reviews have unilaterally targeted static talent areas (i.e. talent detection and identification). This review aimed to identify quantitative and qualitative studies with longitudinal designs, within an elite athlete population, that considered development and selection literature collectively. Taking a novel pragmatic approach achieved pluralism in a strive to greatly advance our methodological understanding to acquire knowledge of more effective talent development in sport. This review followed the Preferred Reporting Items for Systematic Review and used a Meta-aggregation methodology. A search of talent development and selection literature identified 41 quantitative and 3 qualitative longitudinal studies. Overall, ten (quantitative) studies investigated interactions between multidimensional selection (i.e. measures of performance) and development characteristics; performance variables changed non-linearly alongside talent development characteristics. No longitudinal mixed-method research studies were found. For practitioners, multiple performance measures need to be considered alongside development characteristics to better assess talent. For researchers, the design of this review models an epistemological and ontological congruent approach that can be used to facilitate the design of future mixed-method and longitudinal research; capturing the dynamic and multifaceted individual differences of talent development.

Generative Augmentation-Driven Prediction of Diverse Visual Scanpaths in Images

Visual scanpaths of multiple humans on an image represent the process by which they capture the information in it. State-of-the-art models to predict visual scanpaths on images learn directly from recorded human visual scanpaths. However, the generation of multiple visual scanpaths on an image having diversity like human visual scanpaths has not been explicitly considered. In this paper, we propose a deep network for predicting multiple diverse visual scanpaths on an image. Image-specific hidden Markov model based generative data augmentation is performed in the beginning to increase the number of image-visual scanpath training pairs. Considering a similarity between our generative data augmentation process and the use of long short-term memory (LSTM) for prediction, we propose an LSTM based visual scanpath predictor. A network to predict a single visual scanpath on an image is designed first. The network is then modified to predict multiple diverse visual scanpaths representing different viewer varieties by using a parameter indicating the uniqueness of a viewer. A random vector is also employed for subtle variations within scanpaths of the same viewer variety. Our models are evaluated on three standard datasets using multiple performance measures, which demonstrate the superiority of the proposed approach over the state-of-the-art. Empirical studies are also given indicating the significance of our generative data augmentation method and our multiple scanpath prediction strategy producing diverse visual scanpaths. Code link: <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://ashishverma03.github.io/Diverse-Visual-Scanpath</uri>

Dual-process modeling of sequential decision making in the balloon analogue risk task

People are often faced with repeated risky decisions that involve uncertainty. In sequential risk-taking tasks, like the Balloon Analogue Risk Task (BART), the underlying decision process is not yet fully understood. Dual-process theory proposes that human cognition involves two main families of processes, often referred to as System 1 (fast and automatic) and System 2 (slow and conscious). We cross models of the BART with different architectures of the two systems to yield a pool of computational dual-process models that are evaluated on multiple performance measures (e.g., parameter identifiability, model recovery, and predictive accuracy). Results show that the best-performing model configuration assumes the two systems are competitively connected, an evaluation process based on the Scaled Target Learning model of the BART, and an assessment rate that incorporates sensitivity to the trial number, pumping opportunity, and bias to engage in System 1. Findings also shed light on how modeling choices and response times in a dual-process framework can benefit our understanding of sequential risk-taking behavior.

Predictive Modelling of Customer Sustainable Jewelry Purchases Using Machine Learning Algorithms

In order to give businesses useful information, the main goal of this study is to forecast future consumer purchases in the sustainable jewelry sector. Companies launching new goods, services, or modifications to their current offers must anticipate client behaviours. Businesses can learn a lot about how current customers behave by studying their behaviour in order to develop marketing and sales techniques that work. The study uses a variety of statistical and machine learning methods to forecast the most important elements influencing the purchasing of sustainable jewelry in order to achieve this. These variables span a wide range of characteristics, including gender, age, educational attainment, occupation, gross monthly income, marital status, previous jewelry purchase history, purchase intention, preferred shopping locations, online purchase influencers, consumer buying patterns, preferences for branded and non-branded jewelry, and participation in sustainable jewelry practices. In this study, supervised classification algorithms such Naive Bayes, Support Vector Machine, Linear Discriminant Analysis, K-Nearest Neighbor, and Decision Tree are compared over a wide range of machine learning approaches. To increase the predictive model's effectiveness and resilience, ensemble learning techniques including Random Forest, AdaBoost, and Bagging Classifiers are also introduced. A confusion matrix and classification report are the performance indicators used to assess these models. When used on a test dataset, these measures evaluate the model's precision, recall, accuracy, and F1 scores. These measurable metrics are essential for evaluating the models' reliability and efficacy. Notably, the study shows that the approaches of Random Forest, Decision Tree, Bagging Classifier, and Linear Discriminant Analysis continuously produce the highest accuracy rates, all of which stand at a remarkable 100% accuracy. Other used machine learning classifiers, such as Naive Bayes, K-Nearest Neighbor, Support Vector Machine, and AdaBoost algorithms, which provide significantly lower accuracy scores, fall short of this degree of accuracy. The Random Forest Classifier, which boasts a remarkable accuracy score of 100% for the training dataset and a reasonable 53% for the testing dataset, is particularly noteworthy. This classifier surpasses its competitors, demonstrating its potency in forecasting purchases of sustainable jewelry. In order to forecast sustainable jewelry purchases, this study uses a rigorous methodology that draws on a variety of statistical and machine learning models. The Random Forest Classifier is the most accurate and significant model for this predictive task, with Precision, Recall, and F1 scores of 100, according to the results, which were assessed using multiple performance measures like Precision, Recall, and F1 score.

A two-stage iterated greedy algorithm and a multi-objective constructive heuristic for the mixed no-idle flowshop scheduling problem to minimize makespan subject to total completion time

Advanced production systems usually are complex in nature and aim to deal with multiple performance measures simultaneously. Therefore, in most cases, the consideration of a single objective function is not sufficient to properly solve scheduling problems. This paper investigates the multi-objective mixed no-idle flowshop scheduling problem. The addressed optimization case is minimizing makespan subject to an upper bound on total completion time. To solve this problem, we proposed a two-stage iterated greedy and a multi-objective constructive heuristic. Moreover, we developed a new multi-objective improvement procedure focusing on increasing the performance of the developed methods in solving the addressed problem. and a new initialization procedure. We performed several computational tests in order to compare our developed methods with the main algorithms from similar scheduling problems in the literature. It was revealed that the proposed approaches give the best results compared with other state-of-the-art performing methods.

A decreasing failure rate model with a novel approach to enhance the artificial neural network's structure for engineering and disease data analysis

The study focuses on key metrics used to examine the characteristics of a lifetime random variable distribution in reliability and survival theory research. In this analysis, metrics including the probability density function time, mean residual lifespan, mean time between failures, hazard rate, and reliability function are essential. The focus of the inquiry is these important parameters in relation to the Burr-Hatke exponential model specifically. The study focuses on key metrics used to examine the characteristics of a lifetime random variable distribution in reliability and survival theory research. In this analysis, metrics including the probability density function time, mean residual lifespan, mean time between failures, hazard rate, and reliability function are essential The focus of the inquiry is these important parameters in relation to the Burr-Hatke exponential model specifically. A key component of the research is a comparison of the outcomes from the artificial intelligence approach and those from conventional literature-based methodologies. This comparison study sheds light on how well the artificial neural network framework performs while evaluating the Burr-Hatke exponential model's technical features. The study allows a comprehensive analysis of the training and prediction capabilities of the growing neural network by calculating multiple performance measures. This comprehensive strategy improves our comprehension of the model's survival traits and reliability, offering significant contributions to the larger field of study. The network structure's mean square error was estimated to be 5.19E-04, and its coefficient of determination value was 0.99987 for the first neural network model. For the second neural network model, the coefficient of determination value was 0.99999 and the mean square error value was 4.58E-06. The outcomes amply revealed the neural network structure's extraordinarily high prediction accuracy and the degree to which the prediction outputs agree with those of the Maximum Likelihood Estimation technique.

A GRA-based approach for optimal selection of SALBP-1 assembly line balancing heuristics

Assembly line balancing is a compulsory approach which enables mass production of similar parts with proper efficacy. Multiple heuristics are proposed in literature to solve single model assembly line balancing problems (SALBP-1). Previous work has focused on developing and evaluating assembly line balancing heuristics and apply them to different fields. Other work investigates the factors affecting the performance of particular assembly line balancing heuristic. However, there is still a need for a guideline for selecting the most suitable heuristic for a particular application with insights into multiple performance measures. This research presents a grey relational analysis (GRA)-based approach to aid process designers in selecting the heuristic generating that yields the best line balancing. Specifically, combinations of factors of cycle time, order strength, number of tasks, and time variability are considered as operational setups of the line balancing process. The solution of each setup is solved by each heuristic, and the performance of the solution is evaluated. Then, GRA is applied to rank the heuristics on basis of the resulting grey rational grades. The proposed approach provides useful insights into the combinations suitable for a particular heuristic. In this study, time variability and cycle time are shown to have the greatest impact on the overall performance of the considered heuristics.

Following Affirmative and Negated Rules.

Rules are often stated in a negated manner ("no trespassing") rather than in an affirmative manner ("stay in your lane"). Here, we build on classic research on negation processing and, using a finger-tracking design on a touchscreen, we show that following negated rather than affirmative rules is harder as indicated by multiple performance measures. Moreover, our results indicate that practice has a surprisingly limited effect on negated rules, which are implemented more quickly with training, but this effect comes at the expense of reduced efficiency. Only affirmative rules are thus put into action efficiently, highlighting the importance of tailoring how rules are communicated to the peculiarities of the human mind.

Ensemble Learning Based Convolutional Neural Networks for Remaining Useful Life Prediction of Aircraft Engines

Remaining useful life (RUL) prediction is an essential task of Prognostics and Health Management (PHM) of aircraft engines performed utilizing the huge data collected from multiple sensors attached to them to ensure their safe operation. While many studies have been reported on RUL prediction for aircraft engines, only a few of them focus on ensemble learning of CNN models for RUL prediction. The success of ensemble learning which is a combination of several base models developed using either same or different machine learning or deep learning algorithms, critically depends on the diversity among the base models generated. This paper proposes a new data-driven approach for RUL prediction of aircraft engines using ensemble learning based convolution neural networks (CNN) by investigating various steps to generate more diverse base models.&#x0D; The main objectives and contributions of this paper are as follows:&#x0D; &#x0D; Explore various CNN model architectures for RUL prediction - After data preprocessing and exploratory data analysis, two different CNN approaches, namely 2D CNN and 1D CNN with multiple channels, are investigated employing time window approach for time-series input preparation for better feature extraction by CNN. Each approach is experimented with multiple architectures to achieve the best possible outcome.&#x0D; Investigate engine specific RUL target function - For RUL prediction of turbofan engines using the C-MAPSS dataset, typically two RUL target functions, namely linear and piecewise linear, are used to determine RUL target values. In the piecewise linear RUL target function approach, which yields better performance in the reported studies, the RUL target value is assigned based on a piecewise linear degradation model which assumes a constant (and maximum) RUL target value in the early phase before linearly degrading the RUL targets. In the literature, this maximum RUL target value was chosen same for all the engines by taking a value of 125 or 130 without providing proper rationale. In this study, we adopt an approach based on the widely known health index to determine an engine specific initial (and maximum) RUL target value that can be used with the piecewise linear degradation model to determine RUL target values.&#x0D; Investigate hyperparameter optimization of CNN models to generate diverse base models - For the purpose of developing a high performance ensemble CNN model for RUL prediction, hyperparameter optimization of CNN models is performed to optimally determine the network structure (such as # of filters, filter size, stride, padding, # of convolutional, pooling, and dense layers, activation functions etc.) as well as the hyperparameters that determine the network training process (such as optimization method, learning rate, momentum, batch size etc.).&#x0D; Investigate ensemble learning to select and combine diverse CNN models for RUL prediction – In order to develop model combiners, diverse CNN models as base learners are selected using multiple performance measures such as RMSE, score function, MAE, and R2 score, and employing the non-negative least squares method, random forest regression, and extreme learning machine (ELM) to train model combiners.&#x0D; Evaluate the above proposed approach using the C-MAPSS dataset - To show the effectiveness of the proposed approach, various evaluations for RUL prediction using the popular C-MAPSS dataset (including all the four sub-datasets denoted as FD001, FD002, FD003 and FD004) are carried out and the results will be compared against the state-of-the-art results on the same dataset.&#x0D; &#x0D; Major emphasis of this proposed approach is on the generation of diverse CNN base models by carrying out various steps as explained above, and it is expected the results of this proposed approach will contribute towards enhancing the RUL prediction performance especially on the sub-datasets FD002 and FD004 which are challenging for the existing state-of-the-art RUL prediction techniques.

Exploring the impact of missingness on racial disparities in predictive performance of a machine learning model for emergency department triage.

To investigate how missing data in the patient problem list may impact racial disparities in the predictive performance of a machine learning (ML) model for emergency department (ED) triage. Racial disparities may exist in the missingness of EHR data (eg, systematic differences in access, testing, and/or treatment) that can impact model predictions across racialized patient groups. We use an ML model that predicts patients' risk for adverse events to produce triage-level recommendations, patterned after a clinical decision support tool deployed at multiple EDs. We compared the model's predictive performance on sets of observed (problem list data at the point of triage) versus manipulated (updated to the more complete problem list at the end of the encounter) test data. These differences were compared between Black and non-Hispanic White patient groups using multiple performance measures relevant to health equity. There were modest, but significant, changes in predictive performance comparing the observed to manipulated models across both Black and non-Hispanic White patient groups; c-statistic improvement ranged between 0.027 and 0.058. The manipulation produced no between-group differences in c-statistic by race. However, there were small between-group differences in other performance measures, with greater change for non-Hispanic White patients. Problem list missingness impacted model performance for both patient groups, with marginal differences detected by race. Further exploration is needed to examine how missingness may contribute to racial disparities in clinical model predictions across settings. The novel manipulation method demonstrated may aid future research.

Effect of dehydroepiandrosterone therapy on cognitive performance among postmenopausal women: a systematic review of randomized clinical trial data.

Whether dehydroepiandrosterone (DHEA) supplementation improves cognitive performance in older women is uncertain. Nonetheless, DHEA supplements are readily available over the counter in several countries and are potentially being used to prevent cognitive decline and dementia. This systematic review was conducted to evaluate the effect of exogenous DHEA on cognitive performance in postmenopausal women. Ovid MEDLINE, EMBASE, PsycINFO, Web of Science Core Collection, and the Cochrane Central Register of Controlled Trials databases were searched for studies of postmenopausal women until November 30, 2022. Eligible studies were required to be randomized clinical trials, be at least single blind, have a placebo or comparator arm and published in English. Risk of bias of the included studies was assessed by the revised Cochrane risk-of-bias tool. Of the 15 articles retrieved for full-text review, four met the inclusion criteria. In all studies DHEA was administered as a 50-mg oral daily dose and all were double blind with an identical placebo. Three were placebo-controlled, crossover studies and one was a parallel-group clinical trial. The only positive outcome was limited to a 4-wk cross-over study in which DHEA statistically significantly enhanced five of six tests of visual-spatial performance compared with placebo in 24 cognitively normal postmenopausal women. Improvement in cognitive performance with DHEA treatment over placebo group was not seen in any other study. Heterogeneity of design and use of multiple measures of cognitive performance was a barrier to meta-analysis and between study comparisons. The studies were limited by high risk of bias in multiple domains. Overall, this systematic review does not support a beneficial effect of DHEA therapy on cognitive performance in postmenopausal women.

Predicting children’s perceived reading proficiency with prosody modeling

Reading is a foundational skill and the focus of school-level education efforts across countries. The assessment of linguistic competence from oral reading has long been the subject of scientific studies linking the reader’s comprehension of the text to various measures of oral reading fluency. Given the time and resource intensive nature of such assessment, it is of interest to automate the prediction of reading fluency from audio recordings using the same pedagogical rubrics. Given recent findings about the importance of prosody to the communicative purpose of reading aloud, we discuss new approaches to modeling it reliably for the automatic assessment task. We present a new data set of children’s oral reading screened for minimum word decoding skill and rated for comprehensibility by two human experts. We develop a system for the automatic prediction of rater scores that also facilitates insights about the complementarity and inter-dependence of computed lexical accuracy, rate and prosodic features as corroborated by multiple performance measures. With achieved values of correlation and agreement that surpass the corresponding inter-rater measures, we also show how text-dependent prosodic features, informed by speech rate and speaking style, contribute prominently to system performance.

Animal board invited review: Quantification of resilience in farm animals.

Resilience, when defined as the capacity of an animal to respond to short-term environmental challenges and to return to the prechallenge status, is a dynamic and complex trait. Resilient animals can reinforce the capacity of the herd to cope with often fluctuating and unpredictable environmental conditions. The ability of modern technologies to simultaneously record multiple performance measures of individual animals over time is a huge step forward to evaluate the resilience of farm animals. However, resilience is not directly measurable and requires mathematical models with biologically meaningful parameters to obtain quantitative resilience indicators. Furthermore, interpretive models may also be needed to determine the periods of perturbation as perceived by the animal. These applications do not require explicit knowledge of the origin of the perturbations and are developed based on real-time information obtained in the data during and outside the perturbation period. The main objective of this paper was to review and illustrate with examples, different modelling approaches applied to this new generation of data (i.e., with high-frequency recording) to detect and quantify animal responses to perturbations. Case studies were developed to illustrate alternative approaches to real-time and post-treatment of data. In addition, perspectives on the use of hybrid models for better understanding and predicting animal resilience are presented. Quantification of resilience at the individual level makes possible the inclusion of this trait into future breeding programmes. This would allow improvement of the capacity of animals to adapt to a changing environment, and therefore potentially reduce the impact of disease and other environmental stressors on animal welfare. Moreover, such quantification allows the farmer to tailor the management strategy to help individual animals to cope with the perturbation, hence reducing the use of pharmaceuticals, and decreasing the level of pain of the animal.

Behavioral diversity and biomechanical determinants of the outcome of a fish predator–prey interaction

AbstractPredator–prey interactions are ubiquitous and under strong selection because of the consequences experienced by both predator and prey if they lose the interaction. Biomechanics and behavior play important roles in the outcome of these interactions, but many studies focus on the prey, restrict the range of behaviors considered, and the role of prey boldness in the outcome is not understood. We used high‐speed video to test for effects of multiple measures of performance and kinematics of both the predator and prey, and boldness of prey on the outcome of interactions between Pike Cichlids (Crenicichla sp.) and Guppies (Poecilia reticulata). We found high variation in the behaviors employed during the predator–prey interactions, including in suction versus raptorial feeding, strike accuracy, and guppy responsiveness. We also found that predators moving relatively slower and prey moving relatively faster were more successful at consuming the prey and evading the predator, respectively. Prey that reacted farther from the predator was more likely to escape predation, but boldness of the prey did not affect the interaction. Our work suggests that a high level of variation in predator–prey interactions is widespread, even when strike and escape behaviors are stereotyped. We also showed that what both the predator and the prey do during an interaction are important in determining the outcome.

Multistage stochastic fractionated intensity modulated radiation therapy planning

Intensity modulated radiation therapy (IMRT) is a widely used cancer treatment technique designed to target malignant cells. To enhance its effectiveness on tumors and reduce side effects, radiotherapy plans are usually divided into consecutive treatments, or fractions, that are delivered over multiple weeks. However, typical planning approaches have focused on finding the full sequence of radiation intensities prior to the treatment, or were restricted to a single treatment session. In this work, we investigate the fractionated IMRT planning problem that accounts for geometric motion-related uncertainty during treatment. We propose a novel multistage stochastic programming (MSP) modeling framework that incorporates the sequential decision-making nature of the problem and prevailing stochasticity in cancer treatment. We model the uncertainty in the form of a scenario tree following the multiple instance of geometry approximation, and solve the MSP model using stochastic dual dynamic programming considering a variety of risk measures. We conduct computational experiments on five test cases that are generated based on clinical data. Through extensive simulations, we show that our MSP model generates higher quality treatment plans compared to deterministic and two-stage program counterparts based on multiple performance measures. In particular, our model leads to a higher rate of tumor coverage and a lower rate of radiation exposure for healthy tissues. Accordingly, the proposed MSP framework can greatly contribute to the clinical practice in fractionated IMRT.

Multiresponse surface methodology for hyperparameter tuning to optimize multiple performance measures of statistical and machine learning algorithms

AbstractHyperparameter tuning is an important task in machine learning for controlling model complexity and improving prediction performance. Most methods in the literature can only be used to tune hyperparameters to optimize a single performance measure. In practice, participants may want to optimize multiple measures of the model; however, optimizing one measure may worsen another. Therefore, a hyperparameter tuning method is proposed using the multiresponse surface methodology to solve the tradeoff problem between multiple measures. A search algorithm that requires fewer tuning runs is developed to find the optimal hyperparameter settings systematically based on the preferences of participants for multiple measures of the model. An example using the random forest algorithm is provided to demonstrate the application of the proposed method to improve the prediction performance of the model on unbalanced data.

Animal welfare efforts and farm economic outcomes: Evidence from Swedish beef production

AbstractWe estimate the relationship between farm animal welfare (FAW) efforts taken by beef farmers and the economic performance of beef farms by using farm accounting data from the Swedish Farm Economic Survey matched with survey data on farm management practices. To this end, we perform a two-step analysis. First, an item response theory (IRT) model estimates the latent FAW effort on farms. FAW effort likely depends on a host of complementary FAW-improving strategies, and the IRT model combines the considered strategies into a unidimensional scale. We take this to represent on-farm FAW effort. Second, we use instrumental variable regressions to estimate the relationship between FAW effort and multiple measures of farm economic performance. We find that higher FAW effort scores have no effect on margins and costs. However, higher FAW effort scores are associated with lower farm sales. Findings suggest that policies (such as targeted label for high FAW) that increase farm revenue as well as incentivize the uptake of FAW-improvement practices may be able to compensate farmers for their FAW effort.