Selection Model Research Articles

Parental executive functions and motivations unveil variations in young children's screen media use

BackgroundThe increased screen media use among children aged 3 to 5, particularly in the post-COVID era, is concerning. Despite several organizations' recommendation of a one-hour screen limit for young children, actual usage often exceeds this guideline. Objective: This study explored the influence of parental characteristics such as self-efficacy, motivation, socioeconomic status, and cognitive abilities on children's screen time habits.MethodsEmploying a feature selection model, 251 caregivers answered an online survey, presenting data from themselves and on-screen usage for 126 girls and 125 boys. We found that the caregiver’s executive functions, including cognitive flexibility, initiation, task monitoring, and material organization, significantly impact children’s screen time. Results: Our analysis highlighted the vital role of caregivers’ self-efficacy in moderating children's screen usage. Family net income, children's age and gender, and motivations related to children's desires and behavioral control were also significant contributors to usage patterns.ConclusionThis study offers insights into interventions and effective parenting strategies in the digital age, highlighting the importance of addressing socio-demographic factors in understanding this complex issue.

Multi-agent deep reinforcement learning for hyperspectral band selection with hybrid teacher guide

Due to the presence of noisy and highly redundant bands in hyperspectral images (HSIs), band selection serves as a key preprocessing for downstream classification tasks. Recently, deep reinforcement learning (DRL) has been developed as a new trend for band selection of HSIs. Existing DRL-based methods often adopt single-agent, which are prone to fall into local optima due to an excessive action space. The multi-agent methods provide a feasible solution, but often require too much computation. To address these problems, a novel multi-agent DRL method with hybrid teacher guide (MH-DRL) is proposed for band selection of HSIs. In MH-DRL, each agent corresponding to a spectral band decides whether this band is selected. Moreover, a presentation-evaluation network (PE-Net) is constructed to design the reward by evaluating the candidate band subsets without any fine-tuning and represent the state by extracting the spatial-spectral features of HSIs. Then, three kinds of experienced band selection models are regarded as the teachers and designed to participate in the band exploration of DRL, which can improve the learning effectiveness and efficiency by accumulating the external knowledge from diverse teacher models. Finally, deep Q-learning algorithm is designed to update the agents and improve their self-learning ability from continuous exploration. Experimental results on three widely-used HSI data verify the performance of the proposed method better than some advanced band selection algorithms of HSIs. Our source code is available at https://github.com/jiefeng0109/MH-DRL.

Feature Selection for Data Classification in the Semiconductor Industry by a Hybrid of Simplified Swarm Optimization

In the semiconductor manufacturing industry, achieving high yields constitutes one of the pivotal factors for sustaining market competitiveness. When confronting the substantial volume of high-dimensional, non-linear, and imbalanced data generated during semiconductor manufacturing processes, it becomes imperative to transcend traditional approaches and incorporate machine learning methodologies. By employing non-linear classification models, one can achieve more real-time anomaly detection, subsequently facilitating a deeper analysis of the fundamental causes behind anomalies. Given the considerable dimensionality of production line data in semiconductor manufacturing, there arises a necessity for dimensionality reduction to mitigate noise and reduce computational costs within the data. Feature selection stands out as one of the primary methodologies for achieving data dimensionality reduction. Utilizing wrapper-based heuristics algorithms, although characterized by high time complexity, often yields favorable performance in specific cases. If further combined into hybrid methodologies, they can concurrently satisfy data quality and computational cost considerations. Accordingly, this study proposes a two-stage feature selection model. Initially, redundant features are eliminated using mutual information to reduce the feature space. Subsequently, a Simplified Swarm Optimization algorithm is employed to design a unique fitness function aimed at selecting the optimal feature subset from candidate features. Finally, support vector machines are utilized as the classification model for validation purposes. For practical cases, it is evident that the feature selection method proposed in this study achieves superior classification accuracy with fewer features in the context of wafer anomaly classification problems. Furthermore, its performance on public datasets further substantiates the effectiveness and generalization capability of the proposed approach.

Label generation with consistency on the graph for multi-label feature selection

Multi-label feature selection involves the selection of informative features in high-dimensional data sets based on the relationships among different variables. However, large-scale data sets often contain unknown labels that hold latent information, posing a challenge for discovery. Explicitly uncovering latent information among labels not only helps establish robust mappings between features and labels but also expands the applicable knowledge. This paper introduces a novel feature selection method called Label Generation with Consistency on the Graph for Multi-label Feature Selection (LGCM) to effectively integrate label generation and feature selection processes. The proposed method incorporates a two-way flipping mechanism that leverages the consistency on the global graph to guide label generation for each instance. Additionally, the feature selection model utilizes a shared low-rank feature space to minimize deviation from the label generation, providing feedback to the label generation process. Extensive experiments validate the effectiveness of LGCM in improving state-of-the-art feature selection methods in multi-label learning.

An interval-valued spherical fuzzy CIMAS-WISP group decision-analytic model for blockchain platform selection in digital projects

Digital projects aspiring to reach target audiences are executed through decentralized and trustworthy blockchain platforms (BPs). Once the objectives and target audience of a digital project are defined, the selection of suitable BPs is undertaken. The primary objective of this research is to develop a decision support system that aids in the selection of BPs for transferring digital data and assets. Numerous quantitative parameters determine the performance of BPs, alongside qualitative parameters indicating their performance. In this study, the aim is to determine the performance of BPs based on both qualitative and quantitative parameters. Within this scope, a multi-criteria decision-making approach and interval-valued spherical fuzzy (IVSF) sets are adopted. IVSF sets are utilized to determine expert importance levels. The IVSF-criteria importance assessment (CIMAS) method is introduced for the weighting of criteria. IVSF-CIMAS enables the determination of reliability levels in calculating criterion weights. The IVSF-simple weighted sum product (WISP) method is formulated to obtain the performance ranking of BPs. Thus, in this research, the IVSF-CIMAS-WISP hybrid model is developed, and an algorithm for this novel decision-analytic model is presented. A case study is developed focusing on BP selection for a digital project to demonstrate the applicability of the proposed hybrid model. The robustness of IVSF-CIMAS-WISP is tested through extensive sensitivity analysis scenarios. According to the research results, the applicability of the IVSF-CIMAS-WISP hybrid method is supported and its robustness is confirmed. The research findings provide numerous insights for project managers and practitioners.

Gender-specific aspects of socialisation and risk of cardiovascular disease among community-dwelling older adults: a prospective cohort study using machine learning algorithms and a conventional method

BackgroundGender influences cardiovascular disease (CVD) through norms, social relations, roles and behaviours. This study identified gender-specific aspects of socialisation associated with CVD.MethodsA longitudinal study was conducted, involving 9936 (5,231 women...

Discharging driven energy sharing protocol for V2V communication in vehicular energy networks

With the growing concern over range anxiety among electric vehicle (EV) owners due to limited battery capacity and sparse charging infrastructure, EV-to-EV (V2V) energy sharing emerges as a crucial solution to extend driving range. Leveraging the vehicular energy network (VEN) enabled by dynamic wireless power transfer (DWPT) technology, energy sharing among EVs in motion becomes feasible. However, the effective establishment of communication and identification of suitable V2V energy sharing pairs pose significant challenges, particularly for EVs with discharging demands. To address this challenge, we propose a new discharging driven energy sharing protocol based on vehicular ad-hoc networks (VANETs). Firstly, we present a routing approach for transmitting discharging information through VANETs, considering key factors such as distance, state of charge, number of neighbor vehicles, and vehicle speed. This routing scheme facilitates efficient relay node selection on road segments and intersections, ensuring optimal communication paths. Subsequently, we formulate a charging requester selection model to identify the most suitable requester for energy sharing. This model optimizes individual utility while accounting for the state of charge of the requesters, ensuring a comprehensive and inclusive approach to V2V energy sharing. Finally, we develop an acknowledgment message transmission scheme to ensure the completion of selection acknowledgment between discharging EVs and charging EVs. This scheme includes provisions for recovery in case of forwarding link failures, ensuring robust communication in dynamic vehicular environments. Extensive simulations conducted using network simulator 2 (NS-2) demonstrate the superior performance of the proposed protocol in terms of packet delivery ratio, end-to-end delay, and overall V2V energy sharing efficiency.

Model-based estimates for operant selection.

We present a new methodology to partition different sources of behavior change within a selectionist framework based on the Price equation-the multilevel model of behavioral selection. The multilevel model of behavioral selection provides a theoretical background to describe behavior change in terms of operant selection. Operant selection is formally captured by the covariance-based law of effect and accounts for all changes in individual behavior that involve a covariance between behavior and predictors of evolutionary fitness (e.g., food). In this article, we show how the covariance-based law of effect may be applied to different components of operant behavior (e.g., allocation, speed, and accuracy of responding), thereby providing quantitative estimates for various selection effects affecting behavior change using data from a published learning experiment with pigeons.

Individual variation in habitat selection behavior of Asian elephants in a human-wildland interface

Habitat loss and fragmentation due to accelerated agriculture expansion is a major threat to existing wildlife populations across Asia. The human-wildlife interface mosaic across Asia is varied in terms of juxtaposition and structure, which can strongly influence biodiversity value and impacts on wildlife species. Here we analyzed habitat selection behavior of Asian elephant (Elephas maximus) across three study sites with different agriculture use patterns in Myanmar, a country recognized as a global biodiversity hotspot, including but not limited to, commercial palm oil and sugarcane plantations, and subsistence agriculture. These different agriculture use landscape capture landscape mosaic structure found across Asia. Given elephants exhibit heterogeneous spatial behavior, we fitted individual step selection and resource selection models to gain insight into the diversity of strategies employed at the local (step) and home range (third-order) scale. We used variance partitioning analysis to quantify the explanatory contribution of individual and study sites across both scales. We found that the variation in the resource selection behavior was mainly due to individual differences, and the configuration of agriculture present in an individual’s range was the most influential to its selection behavior. Enhancing understanding of how the level of fragmentation on the landscape relates to agricultural use can serve to help focus conservation efforts. Continued accelerated agriculture expansion is increasing the rate of contact between elephants and humans and, thereby escalating negative human-elephant interactions, often resulting in human and elephant deaths. Gaining a deeper understanding of habitat selection behavior by elephants across the changing landscapes of Asia can help inform management decisions and conservation actions.

A Bayesian hierarchical hidden Markov model for clustering and gene selection: Application to kidney cancer gene expression data.

We introduce a Bayesian approach for biclustering that accounts for the prior functional dependence between genes using hidden Markov models (HMMs). We utilize biological knowledge gathered from gene ontologies and the hidden Markov structure to capture the potential coexpression of neighboring genes. Our interpretable model-based clustering characterized each cluster of samples by three groups of features: overexpressed, underexpressed, and irrelevant features. The proposed methods have been implemented in an R package and are used to analyze both the simulated data and The Cancer Genome Atlas kidney cancerdata.

Advancing machine learning for identifying cardiovascular disease via granular computing

<p>Machine learning in cardiovascular disease has broad applications in healthcare, automatically identifying hidden patterns in vast data without human intervention. Early-stage cardiovascular illness can benefit from machine learning models in drug selection. The integration of granular computing, specifically z-numbers, with machine learning algorithms, is suggested for cardiovascular disease identification. Granular computing enables handling unpredictable and imprecise situations, akin to human cognitive abilities. Machine learning algorithms such as Naïve Bayes, K-Nearest Neighbor (KNN), Random Forest, and Gradient Boosting are commonly used in constructing these models. Experimental findings indicate that incorporating granular computing into machine learning models enhances the ability to represent uncertainty and improves accuracy in cardiovascular disease detection.</p>

Hydraulic analysis for assessing environmental flow selection and ecological model formulation

AbstractEcosystem management depends on transforming qualitative observations (e.g., slow‐moving shallow conditions provide nursery refugia for silvery minnow larvae) into management actions to increase habitat quantity or improve habitat quality. To be effective, decision metrics that are developed for management objectives should be validated with field observations. Model assumptions, precision and parameter importance can be refined by comparing the fidelity of selected parameters computed as habitat quality metrics and the correlation of these metrics to real‐world observations. Validated environmental metrics are more credible for management and can be compatible with ecosystem monitoring and project design processes. In this study, streamflow monitoring data and hydraulic modelling are used to quantify fish habitat extent for 15 years of spring runoff. The spring runoff event coincides with larval maturation to a free‐swimming juvenile phase for the silvery minnow, a critical period in Rio Grande habitat management. Different methods to estimate habitat availability (i.e., hydrology statistics, inundation extents based on hydraulic modelling and areal habitat availability based on different formulations of a habitat suitability index curve) were used to test the efficacy of different metrics relative to species population monitoring. This analysis finds that flow–ecology relationships based on hydraulic modelling or hydrology statistics are both effective and highly correlated to larval production. The investigation shows how seasonal hydrologic characterization and hydraulic discretization have varying levels of correlation with seasonal fish production. This study demonstrates how hydraulic modelling data and hydrologic characterization of riverine environments can be used to validate or develop conceptual ecological models.

Strategic forward-looking nonearnings disclosure and overinvestment

We examine whether tone management in different aspects of forward-looking statements (FLSs) is related to managers' self-serving overinvestments. Using data for U.S.-listed firms between 2003 and 2019, we provide novel evidence that the abnormal tone of nonearnings-related qualitative FLSs' is significantly and positively related to firms' overinvestments but that other aspects of FLSs are insignificant to overinvestments. Moreover, this relation is more substantial in financially unconstrained firms. Our findings reveal the heterogeneous roles of different aspects of FLSs in firms' opportunistic disclosures concerning future overinvestments. Further analyses also indicate that this relationship is more pronounced for firms with less monitoring and managers with greater career concerns. We also employ instrumental variables with a two-stage least-square approach and a Heckman selection model to mitigate the endogeneity issue. Our results are robust after conducting a battery of robustness tests. Overall, our findings provide robust evidence that managers are likely to strategically manipulate nonearnings-related qualitative FLSs to mislead investors’ perception of firms' future fundamentals to achieve self-serving purposes.

Determinants of Water Source Choice for Irrigation in the Arkansas Delta

AbstractThe use and the share of water applied from several irrigation water sources correlate with the irrigation practices in use by the peers of Arkansan farmers. From a sample of producers from an irrigation survey in Arkansas, a bivariate sample selection model accounts for how peer use of numerous irrigation practices affects the use and the share of irrigation that comes from a water source. The bivariate sample selection model controls for the bias in the statistical estimates that occur because producers who volunteer for an irrigation survey are likely to know about and use irrigation more than the population. We find that peer influence operates through multiple irrigation practices, and peer influence through an irrigation practice depends on an irrigator's location and current farm practices. For example, peer use of a tail‐water recovery system and peer use of alternate wetting and drying both increase the probability of surface water use alone.

Can machine learning models predict soil moisture evaporation rates? An investigation via novel feature selection techniques and model comparisons

Can machine learning models predict soil moisture evaporation rates? An investigation via novel feature selection techniques and model comparisons

Exploring the genetic basis of resistance to Neopestalotiopsis species in strawberry.

Aggressive strains of Neopestalotiopsis sp. have recently emerged as devastating pathogens of strawberry (Fragaria × ananassa Duchesne ex Rozier), infecting nearly all plant parts and causing severe outbreaks of leaf spot and fruit rot in Florida and globally. The development of host resistance is imperative due to the absence of fungicides that effectively inhibit Neopestalotiopsis sp. growth on an infected strawberry crop. Here, we analyzed 1578 individuals from the University of Florida's (UF) strawberry breeding program to identify and dissect genetic variation for resistance to Neopestalotiopsis sp. and to explore the feasibility of genomic selection. We found that less than 12% of elite UF germplasm exhibited resistance, with narrow-sense heritability estimates ranging from 0.28 to 0.69. Through genome-wide association studies (GWAS), we identified two loci accounting for 7%-16% of phenotypic variance across four trials and 3 years. Several candidate genes encoding pattern recognition receptors, intra-cellular nucleotide-binding leucine-rich repeats, and downstream components of plant defense pathways co-localized with the Neopestalotiopsis sp. resistance loci. Interestingly, favorable alleles at the largest-effect locus were rare in elite UF material and had previously been unintentionally introduced from an exotic cultivar. The array-based markers and candidate genes described herein provide the foundation for targeting this locus through marker-assisted selection. The predictive abilities of genomic selection models, with and without explicitly modeling peak GWAS markers as fixed effects, ranged between 0.25 and 0.59, suggesting that genomic selection holds promise for enhancing resistance to Neopestalotiopsis sp. in strawberry.

The model of choosing a microcontroller for the environmental monitoring system module

Problem. The development and application of modern digital technologies for environmental monitoring is an extremely relevant issue. Such environmental monitoring systems allow for dynamic monitoring of the current state of the studied territories in order to further analyze changes and make decisions on their conservation and development. To improve the environmental situation, it is necessary to constantly improve environmental monitoring systems by creating local environmental monitoring systems operating in real time. Goal. The aim of the study is to develop a model for selecting a microcontroller for an environmental monitoring module based on many functional and cost criteria, which will reduce the time of information collection for a regional environmental monitoring system. Methodology. To achieve this goal, we analyzed the characteristics of microcontrollers. A model for selecting a microcontroller for an environmental monitoring module has been developed based on many functional and cost criteria. An example of using the microcontroller selection model for the environmental monitoring module is given. The method of multicriteria optimization allows the selection of a microcontroller for the module of the environmental monitoring system by calculating the utility function. Results. The paper considers the main tasks of the environmental monitoring system. A mathematical model for selecting a microcontroller for the module of an environmental monitoring system has been developed. It belongs to the class of multicriteria discrete programming models and, unlike the known ones, takes into account the system requirements for the module of the environmental monitoring system as a whole. Originality. The developed model allows us to make a microcontroller according to many functional and cost criteria under conditions of fuzzy input information. Practical value. The use of the proposed model for selecting a microcontroller for the module of the environmental monitoring system will reduce the time of collecting information for the regional environmental monitoring system.

Conceptual underpinnings: theory selection in strategic management research

PurposeTheories are crucial for addressing research questions and advancing the boundaries of knowledge. However, in the field of strategic management, the existence of diverse schools of thought from various disciplines, including economics, politics, and sociology, poses significant challenges for researchers seeking to develop theories for argumentation and theorization. In this study, we have conceptualized a novel approach to selecting an appropriate theory for addressing specific research questions.Design/methodology/approachThought experiment, disciplined imagination, and a conceptual examination of a diverse set of theories.FindingsBecause the central focus in the field of strategic management revolves around how firms achieve sustainable high performance, a research question should initially clarify the fundamental phenomenological issues it aims to address. Subsequently, the process of problematization should identify the ontological assumptions and premises that establish a connection between the research question and existing theories. Finally, the identification and abstraction of rhetorical concepts derived from these assumptions and premises lead to theory selection criteria, namely connectedness, reliability, parsimoniousness, and falsifiability.Research limitations/implicationsAlthough we believe that our model for theory selection is generalizable to a wide range of management disciplines, we have primarily focused on its application in the field of strategic management. Future work could validate and further explore the applicability and effectiveness of this model in selecting appropriate theories for conceptual development in other domains.Originality/valueWhile many researchers have proposed methods for writing theoretical papers, few have provided suggestions specifically focused on theory selection. This paper stands out as one of the few that not only attempts to address this gap but successfully develops a comprehensive model for theory selection.

Selection leads to false inferences of introgression using popular methods.

Detecting introgression between closely related populations or species is a fundamental objective in evolutionary biology. Existing methods for detecting migration and inferring migration rates from population genetic data often assume a neutral model of evolution. Growing evidence of the pervasive impact of selection on large portions of the genome across diverse taxa suggests that this assumption is unrealistic in most empirical systems. Further, ignoring selection has previously been shown to negatively impact demographic inferences (e.g. of population size histories). However, the impacts of biologically realistic selection on inferences of migration remain poorly explored. Here, we simulate data under models of background selection, selective sweeps, balancing selection, and adaptive introgression. We show that ignoring selection sometimes leads to false inferences of migration in popularly used methods that rely on the site frequency spectrum. Specifically, balancing selection and some models of background selection result in the rejection of isolation-only models in favor of isolation-with-migration models and lead to elevated estimates of migration rates. BPP, a method that analyzes sequence data directly, showed false positives for all conditions at recent divergence times, but balancing selection also led to false positives at medium-divergence times. Our results suggest that such methods may be unreliable in some empirical systems, such that new methods that are robust to selection need to be developed.

An integrated Delphi and Fuzzy AHP model for contractor selection: a case of Addis Ababa Design and Construction Works Bureau

An integrated Delphi and Fuzzy AHP model for contractor selection: a case of Addis Ababa Design and Construction Works Bureau