Selective Model Research Articles

Determinants of arable crop farmers’ decisions to adapt to climate change risks in Nigeria

This study ascertained the influence of farmers’ perceptions of climate change effects and their household characteristics on the choice of adaptation technologies they adopt. The survey relied mainly on institutional and primary data for its analysis. Primary data was obtained from arable crop farmers using a set of structured questionnaires administered in a multi-stage, stratified random sampling manner. Sixty (60) farmers were selected from randomly selected states for the five agroclimatic zones in Nigeria, giving a total sample size of 300 farmers. The collected data was analysed using Heckman’s probit selectivity model. It was found that extension contact, gender of the head of household, temperature and rainfall levels determined the decision to adapt to climate change. The respective Z estimates of these aspects were 5.75 (p < 0.01), 5.30 (p < 0.01), 2.32 (p < 0.01) and -9.50 (p < 0.01). However, the farmers’ perceptions of climate change effects were determined by education and agricultural extension access. The Z values for these were 5.42 (p < 0.01) and 2.86 (p < 0.01) respectively. The researcher recommends urgent measures to help farmers adapt to climate change, such as the establishment of weather stations and building the capacities of farmers, especially women.

Automatic base-model selection for white blood cell image classification using meta-learning

Healthcare has benefited from the implementation of deep-learning models to solve medical image classification tasks. For example, White Blood Cell (WBC) image analysis is used to diagnose different pathologies like leukemia. However, medical datasets are mostly imbalanced, inconsistent, and costly to collect. Hence, it is difficult to select an adequate model to overcome the mentioned drawbacks. Therefore, we propose a novel methodology to automatically select models to solve WBC classification tasks. These tasks contain images collected using different staining methods, microscopes, and cameras. The proposed methodology includes meta- and base-level learnings. At the meta-level, we implemented meta-models based on prior-models to acquire meta-knowledge by solving meta-tasks using the shades of gray color constancy method. To determine the best models to solve new WBC tasks we developed an algorithm that uses the meta-knowledge and the Centered Kernel Alignment metric. Next, a learning rate finder method is employed to adapt the selected models. The adapted models (base-models) are used in an ensemble learning approach achieving accuracy and balanced accuracy scores of 98.29 and 97.69 in the Raabin dataset; 100 in the BCCD dataset; 99.57 and 99.51 in the UACH dataset, respectively. The results in all datasets outperform most of the state-of-the-art models, which demonstrates our methodology’s advantage of automatically selecting the best model to solve WBC tasks. The findings also indicate that our methodology can be extended to other medical image classification tasks where is difficult to select an adequate deep-learning model to solve new tasks with imbalanced, limited, and out-of-distribution data.

The use of Active Learning systems for stimulus selection and response modelling in perception experiments

To study the role of perceptual cues on categorization and decision making, participants are typically tested in (perception) experiments with a fixed set of randomized or pseudo-randomized trials. In linguistics and psycholinguistics, for instance, studies often investigate the relative weighting of different cues for a linguistic contrast (e.g., intonation vs. word order). For categorization beyond the segmental level (e.g., /p/ vs. /b/), it is important to establish that results generalise to different words or sentences, which necessitates the use of a range of different items. This may limit the number of conditions (cues and cue combinations) that can be sensibly tested in the same experiment. We show that Active Learning (AL) systems provide a solution: Since stimulus selection is informed by the system's learning mechanism (presenting certain conditions less often than uncertain conditions), they allow for efficient testing of numerous conditions and different items in the same experiment. In this paper, we compared two weighting approaches (average probability-based vs. regression-based) to model the outcome of three simulated scenarios with three binary factors each. Results show that valid results (i.e., little error between predicted values and the actual responses at the end of the experiment) are obtained after about a third of the trials of an original psycholinguistic experiment we replicated. For simulations with interactions between factors, the regression-based approach performed better. Our findings bear implications for the application of AL in psycholinguistic research (extraction of cue weights, inferential statistics, and a stopping criterion during an on-going experiment), which we will discuss.

Learning to Select Prototypical Parts for Interpretable Sequential Data Modeling

Prototype-based interpretability methods provide intuitive explanations of model prediction by comparing samples to a reference set of memorized exemplars or typical representatives in terms of similarity. In the field of sequential data modeling, similarity calculations of prototypes are usually based on encoded representation vectors. However, due to highly recursive functions, there is usually a non-negligible disparity between the prototype-based explanations and the original input. In this work, we propose a Self-Explaining Selective Model (SESM) that uses a linear combination of prototypical concepts to explain its own predictions. The model employs the idea of case-based reasoning by selecting sub-sequences of the input that mostly activate different concepts as prototypical parts, which users can compare to sub-sequences selected from different example inputs to understand model decisions. For better interpretability, we design multiple constraints including diversity, stability, and locality as training objectives. Extensive experiments in different domains demonstrate that our method exhibits promising interpretability and competitive accuracy.

Investigation of cavitation properties of a mobile pumping unit

BACKGROUND: In the introduction to the article, a review of publications on cavitation, vibration and noise in centrifugal pumps, including the issues of cavitation erosion of impellers, is carried out. AIMS: Comparison of cavitation properties of a centrifugal pump of a mobile pumping unit with and without a pre-engineered screw by computational fluid dynamic (CFD) modeling. METHODS: The calculation of the flow part of a pre-injected impeller stage is described and the CFD model of its hydrodynamic simulation is described. In the CFD model, Navier-Stokes equations averaged over the Reynolds number and the working fluid continuity equation were used. A two-phase fluid model was used to simulate cavitation. RESULTS: The final results of the calculations carried out in the above models are presented. Calculations were obtained for a pump with impeller with and without an upstream stage (screw). For the impeller without a screw, the cavitation margin of 4.7 m was obtained, which is critical for such a pump. For a pump with an impeller with an upstream auger the cavitation margin is 1,7 m, that is much better and allows to show efficiency of such solution. CONCLUSIONS: The requirement of hydrodynamic modeling for selection of optimal flow part of centrifugal pump to improve its cavitation characteristics is formulated.

Prediction of posttraumatic functional recovery in middle-aged and older patients through dynamic ensemble selection modeling.

Age-specific risk factors may delay posttraumatic functional recovery; complex interactions exist between these factors. In this study, we investigated the prediction ability of machine learning models for posttraumatic (6 months) functional recovery in middle-aged and older patients on the basis of their preexisting health conditions. Data obtained from injured patients aged ≥45 years were divided into training-validation (n = 368) and test (n = 159) data sets. The input features were the sociodemographic characteristics and baseline health conditions of the patients. The output feature was functional status 6 months after injury; this was assessed using the Barthel Index (BI). On the basis of their BI scores, the patients were categorized into functionally independent (BI >60) and functionally dependent (BI ≤60) groups. The permutation feature importance method was used for feature selection. Six algorithms were validated through cross-validation with hyperparameter optimization. The algorithms exhibiting satisfactory performance were subjected to bagging to construct stacking, voting, and dynamic ensemble selection models. The best model was evaluated on the test data set. Partial dependence (PD) and individual conditional expectation (ICE) plots were created. In total, nineteen of twenty-seven features were selected. Logistic regression, linear discrimination analysis, and Gaussian Naive Bayes algorithms exhibited satisfactory performances and were, therefore, used to construct ensemble models. The k-Nearest Oracle Elimination model outperformed the other models when evaluated on the training-validation data set (sensitivity: 0.732, 95% CI: 0.702-0.761; specificity: 0.813, 95% CI: 0.805-0.822); it exhibited compatible performance on the test data set (sensitivity: 0.779, 95% CI: 0.559-0.950; specificity: 0.859, 95% CI: 0.799-0.912). The PD and ICE plots showed consistent patterns with practical tendencies. Preexisting health conditions can predict long-term functional outcomes in injured middle-aged and older patients, thus predicting prognosis and facilitating clinical decision-making.

Demographic history and distinct selection signatures of two domestication genes in mungbean.

Domestication is the long and complex process underlying the evolution of crops, in which artificial directional selection transformed wild progenitors into the desired form, affecting genomic variation and leaving traces of selection at targeted loci. However, whether genes controlling important domestication traits follow the same evolutionary pattern expected under the standard selective sweep model remains unclear. With whole-genome re-sequencing of mungbean (Vigna radiata), we investigated this issue by resolving its global demographic history and targeted dissection of the molecular footprints of genes underlying two key traits representing different stages of domestication. Mungbean originated in Asia, and the Southeast Asian wild population migrated to Australia about 50 thousand generations ago (kga). Later in Asia, the cultivated form diverged from the wild progenitor. We identified the gene associated with the pod shattering resistance trait, VrMYB26a, with lower expression across cultivars and reduced polymorphism in the promoter region, reflecting a hard selective sweep. On the other hand, the stem determinacy trait was associated with VrDet1. We found two ancient haplotypes of this gene have lower gene expression and exhibited intermediate frequencies in cultivars, consistent with selection favoring independent haplotypes in a soft selective sweep. In mungbean, contrasting signatures of selection were identified from the detailed dissection of two important domestication traits. The results suggest complex genetic architecture underlying the seemingly simple process of directional artificial selection and highlight the limitations of genome-scan methods relying on hard selective sweeps.

The influence of social comparison on risk decision-making for self and groups in intergroup contexts

ABSTRACT Objective The current study aims to explore the influence of social comparison on risk decision-making for self and for groups in intergroup contexts. Method Two experiments with the within-subjects design of 2 (social comparison: upward comparison, downward comparison) × 3 (decision-maker role: for “me”, for “us”, for “them”) were conducted in this study. Experiment 1 focused on the ingroup contexts, and experiment 2 focused on the outgroup contexts. Results (1) in outgroup contexts, individuals are more risk-seeking in upward comparison conditions than in downward comparison conditions. However, the difference disappears in ingroup contexts. (2) Making decisions for “them” is riskier than making decisions for “me” and for “us” with no significant differences between the latter two and consistent across intergroup contexts. (3) The difference in risk decisions made amid upward and downward comparisons is amplified for decisions made for groups. Conclusion The findings may support the selective accessibility model and provide an interpretation with responsibility alleviation for self-group differences in risk decision-making.

Multi-scale habitat selection modeling using combinatorial optimization of environmental covariates: A case study on nature reserve of red-crowned cranes

Multi-scale habitat selection modeling (HSM) has garnered attention due to its ability to incorporate scale dependence of species. The key of multi-scale HSM is to select the appropriate combination of scales for different resources or environmental conditions, and then construct a set of multi-scale environmental covariates as the features of HSM. However, the existing scale selection methods do not determine the combination of scales under a unified model. In this study, a combinatorial optimization approach is proposed. We regard the combination of different scales as a search space, and use a heuristic optimization algorithm to search for the best-fitting model to determine the optimal scales for each resource or environmental condition. In a case study conducted in Yancheng National Nature Reserve, the proposed approach is applied to model the habitat selection of the endangered red-crowned crane. We compare the proposed method with single-scale, random-scale and other multi-scale approaches. The results show that the combination of scales selected based on the proposed method obtained the best accuracy in the spatial prediction of habitat suitability with a test AUC of 0.865 for the daytime scenario and 0.932 for the nighttime scenario. Moreover, the selected scales are utilized to generate response curves, providing suggestions for habitat restoration and management of the red-crowned crane population in the nature reserve.

Non-surgical Aesthetic Treatment Conversion to Surgery: Implications for Patient Selection and Practice Modeling

Non-surgical Aesthetic Treatment Conversion to Surgery: Implications for Patient Selection and Practice Modeling

Holographic dissipation prefers the Landau over the Keldysh form

Although holographic duality has been regarded as a complementary tool in helping understand the nonequilibrium dynamics of strongly coupled many-body systems, it still remains a remarkable challenge how to confront its predictions quantitatively with real experimental scenarios. By matching the holographic vortex dynamics with the phenomenological dissipative Gross-Pitaevskii models, we find that the holographic dissipation mechanism can be well captured by the Landau form rather than the Keldysh one, although the latter is much more widely used in numerical simulations. Our finding is expected to open up novel avenues for facilitating the quantitative test of holographic predictions against upcoming experimental data. Our result also provides a prime example how holographic duality can help select proper phenomenological models to describe far-from-equilibrium nonlinear dynamics beyond the hydrodynamic regime.

A phase IIb, open-label, randomized controlled dose ranging multi-centre trial to evaluate the safety, tolerability, pharmacokinetics and exposure-response relationship of different doses of delpazolid in combination with bedaquiline delamanid moxifloxacin in adult subjects with newly diagnosed, uncomplicated, smear-positive, drug-sensitive pulmonary tuberculosis

BackgroundLinezolid is an effective, but toxic anti-tuberculosis drug that is currently recommended for the treatment of drug-resistant tuberculosis. Improved oxazolidinones should have a better safety profile, while preserving efficacy. Delpazolid is a novel oxazolidinone developed by LegoChem Biosciences Inc. that has been evaluated up to phase 2a clinical trials. Since oxazolidinone toxicity can occur late in treatment, LegoChem Biosciences Inc. and the PanACEA Consortium designed DECODE to be an innovative dose-ranging study with long-term follow-up for determining the exposure–response and exposure–toxicity relationship of delpazolid to support dose selection for later studies. Delpazolid is administered in combination with bedaquiline, delamanid and moxifloxacin.MethodsSeventy-five participants with drug-sensitive, pulmonary tuberculosis will receive bedaquiline, delamanid and moxifloxacin, and will be randomized to delpazolid dosages of 0 mg, 400 mg, 800 mg, 1200 mg once daily, or 800 mg twice daily, for 16 weeks. The primary efficacy endpoint will be the rate of decline of bacterial load on treatment, measured by MGIT liquid culture time to detection from weekly sputum cultures. The primary safety endpoint will be the proportion of oxazolidinone class toxicities; neuropathy, myelosuppression, or tyramine pressor response.Participants who convert to negative liquid media culture by week 8 will stop treatment after the end of their 16-week course and will be observed for relapse until week 52. Participants who do not convert to negative culture will receive continuation phase treatment with rifampicin and isoniazid to complete a six-month treatment course.DiscussionDECODE is an innovative dose-finding trial, designed to support exposure-response modelling for safe and effective dose selection. The trial design allows assessment of occurrence of late toxicities as observed with linezolid, which is necessary in clinical evaluation of novel oxazolidinones. The primary efficacy endpoint is the change in bacterial load, an endpoint conventionally used in shorter dose-finding trials. Long-term follow-up after shortened treatment is possible through a safety rule excluding slow-and non-responders from potentially poorly performing dosages.Trial registrationDECODE was registered in ClinicalTrials.gov before recruitment start on 22 October 2021 (NCT04550832).

Population genetics: coalescence rate and demographic parameters inference

Inferring the demographic history of species is a great challenge in population genetics. This history is classically represented as a history of size changes, ignoring population structure. We present here the work carried out over the last decade around the concept of IICR (Inverse Instantaneous Coalescence Rate), which makes it possible to link, on the one hand, the history of the true population size for a panmictic population, and the inferred size, sometimes called "effective size", when structure is taken into account. We show that population structure can lead to misinterpretations of some demographic history inference results, we propose a framework for inferring structure-specific demographic parameters (number and size of subpopulations, migration rates), and we analyze the link between IICR and some form of selection modeling on genetic sequences.

Advances in tropical cyclone prediction on subseasonal time scales during 2019–2022

This review describes advances in understanding and forecasting tropical cyclone (TC) subseasonal variability during the past four years. A large effort by the scientific community has been in understanding the sources of predictability at subseasonal timescales beyond the well-known modulation of TC activity by the Madden-Julian Oscillation (MJO). In particular, the strong modulation of TC activity over the western North Pacific by the Boreal Summer Intra-Seasonal Oscillation (BSISO) has been documented. Progress has also been realized in understanding the role of tropical-extratropical interactions in improving subseasonal forecasts. In addition, several recent publications have shown that extratropical wave breaking may have a role in the genesis and development of TCs. Analyses of multi-model ensemble data sets such as the Subseasonal to Seasonal (S2S) and Subseasonal Experiment (SubX) have shown that the skill of S2S models in predicting the genesis of TCs varies strongly among models and regions but is often tied to their ability to simulate the MJO and its impacts. The skill in select models has led to an increase over the past four years in the number of forecasting centers issuing subseasonal TC forecasts using various techniques (statistical, statistical-dynamical and dynamical). More extensive verification studies have been published over the last four years, but often only for the North Atlantic and eastern North Pacific.

A study on the mesh size selectivity by a lternate haul method of trawl using the SELECT model

A study on the mesh size selectivity by a lternate haul method of trawl using the SELECT model

Ex vivo to in vivo model of malignant peripheral nerve sheath tumors for precision oncology.

Malignant peripheral nerve sheath tumors (MPNST) are aggressive soft tissue sarcomas that often develop in patients with neurofibromatosis type 1 (NF1). To address the critical need for novel therapeutics in MPNST, we aimed to establish an ex vivo 3D platform that accurately captured the genomic diversity of MPNST and could be utilized in a medium-throughput manner for drug screening studies to be validated in vivo using patient-derived xenografts (PDX). Genomic analysis was performed on all PDX-tumor pairs. Selected PDX were harvested for assembly into 3D microtissues. Based on prior work in our labs, we evaluated drugs (trabectedin, olaparib, and mirdametinib) ex vivo and in vivo. For 3D microtissue studies, cell viability was the endpoint as assessed by Zeiss Axio Observer. For PDX drug studies, tumor volume was measured twice weekly. Bulk RNA sequencing was performed to identify pathways enriched in cells. We developed 13 NF1-associated MPNST-PDX and identified mutations or structural abnormalities in NF1 (100%), SUZ12 (85%), EED (15%), TP53 (15%), CDKN2A (85%), and chromosome 8 gain (77%). We successfully assembled PDX into 3D microtissues, categorized as robust (>90% viability at 48 h), good (>50%), or unusable (<50%). We evaluated drug response to "robust" or "good" microtissues, namely MN-2, JH-2-002, JH-2-079-c, and WU-225. Drug response ex vivo predicted drug response in vivo, and enhanced drug effects were observed in select models. These data support the successful establishment of a novel 3D platform for drug discovery and MPNST biology exploration in a system representative of the human condition.

Research on the Impact of CFO Background Heterogeneity on IPO Under-Pricing

In the current market, IPO underpricing is widespread, which will bring a series of serious consequences to the company and the whole market. In the enterprise, the chief financial officer (CFO), as the financial core of the senior management, bears incomparable responsibilities for the enterprise finance. The role of the chief financial officer can be further reflected in the IPO process of the company. Generally speaking, the CFOs in different stages of listing are different, so their impact on IPO underpricing is also different. Based on this background, this paper takes the micro perspective of CFO's personal characteristics as the starting point to explore its impact on IPO underpricing rate. Firstly, this paper summarizes and introduces the relevant domestic and foreign research on the influencing factors of IPO Underpricing and the background characteristics of CFO. On the basis of understanding the research conclusions made by existing scholars, this paper considers the current market situation, studies relevant theories, considers the expected purpose, and puts forward a series of hypotheses combined with the existing results. Secondly, select effective indicators and models for empirical analysis to test the impact of the background characteristics of the CFO on the IPO underpricing range, and analyze whether the gender characteristics, educational characteristics and financial professional background characteristics of the CFO have a significant impact on the IPO underpricing range. Finally, according to the research results, this paper puts forward relevant suggestions to reduce IPO Underpricing and summarizes it.

Cognitive factors in Thai-naïve Mandarin and Vietnamese speakers’ imitation of Thai lexical tones

The present study examined how native phonological and phonetic factors in non-native speech perception (Perceptual Assimilation Model [PAM]: Best, 1995) affect non-native imitation of Thai tones by Thai-naïve Mandarin and Vietnamese participants, and how memory load and stimulus variability shift the balance between phonological versus phonetic modes (Automatic Selective Perception model [ASP]: Strange, 2011) in imitation. Although overall imitation was quite good, native phonological influences, as reflected in Categorised versus Uncategorised assimilation types in a prior perception study with the same participants (Chen, Antoniou, &amp;amp; Best, 2023), constrained non-native tone imitations. Residual perceptual sensitivity to phonetic differences between the target stimuli and native tones, as reflected in percent choice and goodness ratings in assimilation, also affected imitation. Effects of stimulus variability and memory load were restricted and interacted with specific tones for each participant group. Imitations were generally more accurate under low memory load, constant talker and vowel conditions, consistent with a more phonetic mode of perception. Imitation accuracy decreased under high memory load, variable talker and vowel conditions, consistent with a more phonological mode of perception. Results support the PAM and ASP accounts of native phonological and phonetic effects in non-native perception and extend them to non-native tone imitation.

A novel multimodal online news popularity prediction model based on ensemble learning

AbstractThe prediction of news popularity is having substantial importance for the digital advertisement community in terms of selecting and engaging users. Traditional approaches are based on empirical data collected through surveys and applied statistical measures to prove a hypothesis. However, predicting news popularity based on statistical measures applied to past data is highly questionable. Therefore, in this article, we predict news popularity using machine learning classification models and deep residual neural network models. Articles are usually made up of textual content and in many cases, images are also used. Although it is evident that the appropriate amount of textual data is required to extract features and create models, image data is also helpful in gaining useful information. In this article, we present a novel multimodal online news popularity prediction model based on ensemble learning. This research work acts as a guide for extensive feature engineering, feature extraction, feature selection and effective modelling to create a robust news popularity Prediction Model. Three kinds of features—meta‐features, text features and image features are used to design an influential and robust model. The relative error performance measure Root Mean Squared logarithmic error (RMSLE) is used to quantify the popularity prediction error. Further, the RMSLE outcome shows 0.351 which is the lowest error value given by the proposed model. Further, the most important features are also sought out to show the dependence of the best‐fit model on text and image features.

Deep Proteomics Network and Machine Learning Analysis of Human Cerebrospinal Fluid in Japanese Encephalitis Virus Infection.

Japanese encephalitis virus is a leading cause of neurological infection in the Asia-Pacific region with no means of detection in more remote areas. We aimed to test the hypothesis of a Japanese encephalitis (JE) protein signature in human cerebrospinal fluid (CSF) that could be harnessed in a rapid diagnostic test (RDT), contribute to understanding the host response and predict outcome during infection. Liquid chromatography and tandem mass spectrometry (LC-MS/MS), using extensive offline fractionation and tandem mass tag labeling (TMT), enabled comparison of the deep CSF proteome in JE vs other confirmed neurological infections (non-JE). Verification was performed using data-independent acquisition (DIA) LC-MS/MS. 5,070 proteins were identified, including 4,805 human proteins and 265 pathogen proteins. Feature selection and predictive modeling using TMT analysis of 147 patient samples enabled the development of a nine-protein JE diagnostic signature. This was tested using DIA analysis of an independent group of 16 patient samples, demonstrating 82% accuracy. Ultimately, validation in a larger group of patients and different locations could help refine the list to 2-3 proteins for an RDT. The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE partner repository with the dataset identifier PXD034789 and 10.6019/PXD034789.