Skewness Research Articles

TSLiNGAM: DirectLiNGAM Under Heavy Tails

One of the established approaches to causal discovery consists of combining directed acyclic graphs (DAGs) with structural causal models (SCMs) to describe the functional dependencies of effects on their causes. Possible identifiability of SCMs given data depends on assumptions made on the noise variables and the functional classes in the SCM. For instance, in the LiNGAM model, the functional class is restricted to linear functions and the disturbances have to be non-Gaussian. In this work, we propose TSLiNGAM, a new method for identifying the DAG of a causal model based on observational data. TSLiNGAM builds on DirectLiNGAM, a popular algorithm which uses simple OLS regression for identifying causal directions between variables. TSLiNGAM leverages the non-Gaussianity assumption of the error terms in the LiNGAM model to obtain more efficient and robust estimation of the causal structure. TSLiNGAM is justified theoretically and is studied empirically in an extensive simulation study. It performs significantly better on heavy-tailed and skewed data and demonstrates a high small-sample efficiency. In addition, TSLiNGAM also shows better robustness properties as it is more resilient to contamination. Supplementary materials for this article are available online.

Thin corridors limit wildlife: Variance of tropical carnivore distribution and habitat use in a critical rainforest corridor

AbstractMadagascar is acclaimed for its floral and faunal endemism and biodiversity. Among the island nation's most emblematic fauna are its native mammalian carnivores; they are members of the threatened and endemic Eupleridae family. The Corridor of Marojejy, Anjanaharibe‐Sud Special Reserve, Tsaratanana (COMATSA) is a newly protected corridor system that faces deforestation and lacks detailed assessments of its native carnivore community. We deployed 44 motion‐activated trail cameras to identify which terrestrial carnivores are found within the rainforest corridor landscape of COMATSA‐Sud and Marojejy National Park, and to examine carnivore occupancy, relative activity and the impacts of habitat edge. Our sampling array operated from October 20, 2021, to February 10, 2022 (113 days) and confirmed the presence of four species in Eupleridae: Cryptoprocta ferox, Fossa fossana, Galidia elegans, and Galidictis fasciata, as well as a non‐native carnivore species: Canis familiaris. We show that carnivore occurrences significantly increase with distance to forest edges. We discuss the negative impacts of forest loss on native carnivores and highlight the threats posed by the non‐native carnivore C. familiaris. Our findings spotlight faunal detections across the corridor and indicate that fewer native carnivores are found in the corridor compared to nearby protected areas, suggesting that deforestation correlates with a skewed distribution of carnivore detections. Based on our data, we provide management recommendations for the protection of the forest corridor and its endemic faunal community.

Normalization of timed measures in bilingualism research

Abstract The time it takes an individual to respond to a probe (e.g., a word, picture, or question) or to read a word or phrase provides useful insights into cognitive processes. Consequently, timed measures are a staple in bilingualism research. However, timed measures usually violate assumptions of linear models, one being normal distribution of the residuals. Power transformations are a common solution but which of the many possible transformations to apply is often guesswork. Box and Cox (1964) developed a procedure to estimate the best-fitting normalizing transformation, coefficient lambda (λ), that is easy to run using standard R packages. This practical primer demonstrates how to perform the Box-Cox transformation in R using as a testbed the distractor items from a recent eye-tracking study on sentence reading in speakers of Spanish as a majority and a heritage language. The analyses show (a) that the exponents selected via the Box-Cox procedure reduce positive skewness as well as or better than the natural log; (b) that the best-fitting value of λ varies based on factors such as group and, in the case of eye-movement data, the measure of interest; and (c) that the choice of transformation sometimes impacts p values for model estimates.

Multifractal structure and Gutenberg–Richter parameter associated with volcanic emissions of high energy in Colima, Mexico (years 2013–2015)

Abstract. The evolution of multifractal structures in various physical processes, such as climatology, seismology, or volcanology, serves as a crucial tool for detecting changes in corresponding phenomena. In this study, we explore the evolution of the multifractal structure of volcanic emissions with varying energy levels (observed at Colima, Mexico, during the years 2013–2015) to identify clear indicators of imminent high-energy emissions nearing 8.0×108 J. These indicators manifest through the evolution of six multifractal parameters: the central Hölder exponent (α0); the maximum and minimum Hölder exponents (αmax, αmin); the multifractal amplitude (W=αmax-αmin); the multifractal asymmetry (γ=[αmax-α0]/[α0-αmin]); and the complexity index (CI), calculated as the sum of the normalized values of α0, W, and γ. Additionally, the results obtained from adapting the Gutenberg–Richter seismic law to volcanic energy emissions, along with the corresponding skewness and standard deviation of the volcanic emission data, further support the findings obtained through multifractal analysis. These results, derived from multifractal structure analysis, adaptation of the Gutenberg–Richter law to volcanic emissions, and basic statistical parameters, hold significant relevance in anticipating potential volcanic episodes of high energy. Such anticipation can be further quantified using an appropriate forecasting algorithm.

Digital Transformation and Green Innovation Synergy: Double Drivers for Global Value Chain Upgrading in China's Construction Machinery Industry

With the digital transformation of the global economy and the deepening of the concept of sustainable development, China's construction machinery industry faces major opportunities and challenges in the reconstruction and upgrading of the global value chain. Based on the dynamic capability theory and the collaborative innovation theory, this study constructs a theoretical framework of "digital transformation-green innovation-global value chain upgrading" to explore the impact of the synergy between digital transformation and green innovation on improving the position of Chinese construction machinery companies in the global value chain. This study uses a multi-source data fusion method, combined with patent data, financial statements and questionnaires, to construct an unbalanced panel data set of 189 listed construction machinery companies in China from 2010 to 2022, and uses econometric methods such as dynamic panel modeling, threshold regression and machine learning to analyze. The study finds that: (1) The impact of digital transformation on the global value chain upgrading of construction machinery companies presents a nonlinear "U-shaped" curve relationship; (2) Green innovation plays a positive regulatory role in the relationship between digital transformation and value chain upgrading; (3) There is a significant synergy between digital transformation and green innovation ; (4) The quality of the institutional environment has an important impact on the synergy, and a high-quality institutional environment enhances the synergy; (5) Enterprise heterogeneity significantly moderates the impact of synergy on value chain upgrading. The study expands the theory of global value chains, enriches the theory of collaborative innovation, provides a new theoretical perspective for the integration of digital economy and green development, and provides practical guidance for enterprises and governments to formulate relevant strategies and policies. Future research can further explore the cross-national comparison and long-term dynamic evolution of synergy effects.Digital transformation

Robust variable selection methods with Cox model-a selective practical benchmark study.

With the advancement of biological and medical techniques, we can now obtain large amounts of high-dimensional omics data with censored survival information. This presents challenges in method development across various domains, particularly in variable selection. Given the inherently skewed distribution of the survival time outcome variable, robust variable selection methods offer potential solutions. Recently, there has been a focus on extending robust variable selection methods from linear regression models to survival models. However, despite these developments, robust methods are currently rarely used in practical applications, possibly due to a limited appreciation of their overall good performance. To address this gap, we conduct a selective review comparing the variable selection performance of twelve robust and non-robust penalised Cox models. Our study reveals the intricate relationship among covariates, survival outcomes, and modeling approaches, demonstrating how subtle variations can significantly impact the performance of methods considered. Based on our empirical research, we recommend the use of robust Cox models for variable selection in practice based on their superior performance in presence of outliers while maintaining good efficiency and accuracy when there are no outliers. This study provides valuable insights for method development and application, contributing to a better understanding of the relationship between correlated covariates and censored outcomes.

Chile Niño/Niña in the coupled model intercomparison project phases 5 and 6

The north and central coast of Chile is influenced by El Niño-Southern Oscillation (ENSO) through oceanic and atmospheric teleconnections. However, it also experiences episodic oceanic warmings off central Chile (30°S) lasting a few months that are not necessarily associated with ENSO. These episodes, called “Chile Niño” events, besides their ecological and socio-economical impacts, have also the potential to influence tropical Pacific variability. Here, we investigate how realistically the models in the Coupled Model Intercomparison Project (CMIP, Phases 5 and 6) simulate Chile Niño/Niña (CN) events, and quantify their changes under anthropogenic forcing. Despite limitations of the global models in simulating realistically coastal upwelling dynamics, we show that they simulate reasonably well the observed spatial pattern, amplitude and seasonal evolution of CN events. They however fail to properly represent the positive skewness from observations. The analysis of a sub-group of models (36) that simulate ENSO realistically reveals that CN events increase in amplitude and variance in the future climate with no changes in their frequency of occurence. This is interpreted as resulting from compensating effects amongst changes in remote drivers and local feedbacks. In particular, ENSO variance increases while that of the South Pacific Oscillation decreases. Conversely, we found that while the Wind-Evaporation-SST feedback tends to increase and the coupling between mixed-layer depth and SST weakens, favoring the development of CN events, the thermocline and wind-SST feedbacks decrease. However, only the change in the thermocline feedback is correlated to changes in CN variance amongst the models, suggesting a dominant role of local oceanic stratification changes in constraining the sensitivity of CN to global warming.

Hydrodynamic Analysis of Sediment Deposition in Mangrove Wetlands of Zhanjiang Bay-Based on the Perspective of Sediment Grain Size

Grain size is an important parameter reflecting the characteristics of sedimentary dynamics, and mangrove wetland sediments record abundant sedimentary environmental information during transportation and deposition. To investigate the vertical characteristics of sediment grain size and sedimentary dynamic processes in the mangrove wetlands of Zhanjiang Bay, this study analyzed the grain size parameters, frequency distribution curves, and hydrodynamic characteristics of sediment cores collected from the mangrove wetlands in Nansan Island (A), the south of Techeng Island (B), Guanhai Promenade (C), Dongtoushan Island (D), Huguang Town (E), and the west of Donghai Island (F). The objective was to reveal the evolution of sedimentary dynamics in Zhanjiang Bay over the past century. The results showed that: (1) There were significant differences in the grain size characteristics among the sediment cores. Cores A and B had poor sorting and flat kurtosis, with sediment types of silty sand. Core C had a sinuous mean grain size and kurtosis curve, poor sorting, positive skewness, and a sediment type of silty sand. Core D had negative skewness, poor sorting, and moderate kurtosis, with a sediment type of sandy silt. Cores E and F had negative skewness, poor sorting, and large variations in kurtosis, with sediment types of sandy silt; (2) The grain size frequency distribution curves at depths of 25 cm, 50 cm, 75 cm, and 100 cm in each sediment core exhibited bimodal and unimodal characteristics, indicating the involvement of multiple sedimentary media in the depositional process. Cores A, B, C, and D had grain sizes concentrated on the sandy particle end, indicating strong sedimentary dynamics. Cores E and F had prominent silt-sized grains, indicating weaker sedimentary dynamics; (3) The grain size characteristics of the sediments were related to their terrestrial or island origins and the sediment supply from runoff. The depositional processes were deeply influenced by tidal dynamics, mangrove distribution, and human activities. The accelerated deposition rate in Zhanjiang Bay in recent years may affect navigation channels, and thus, attention should be paid to the ecological environment of wetlands within the bay.

Influence of green practices on user loyalty in sport clubs

The sports industry contributes significantly to environmental degradation through its economic practices. As a result, there has been interest in reducing the impact through various initiatives in sectors such as sporting events, facility management, etc. However, these initiatives have not been studied in the sport sector, eluding the beneficial effects it could have on market positioning and user retention. Therefore, this study aims to assess the effect of green practices implemented by sports clubs on their users’ perceptions and loyalty. The GPSport and EPOD2 questionnaires were used on a total of 1732 users of 27 Spanish sports clubs, of whom 70% were men. In data processing, standard deviation, skewness, and kurtosis were used for psychometric properties; Cronbach’s alpha, AVE, and CR were used for correlations between factors, and CMIN, DF, and CFI were used to test the invariance of the factor structure across user groups. Perceptions of the service were predominantly positive. The construct with the highest value was satisfaction (4.36 ± 0.73), followed by loyalty (4.19 ± 0.79), with green practices scoring the lowest (3.01 ± 0.94). Perceived quality was a significant and direct antecedent of satisfaction, while satisfaction was a significant and direct antecedent of loyalty, with the former relationship being the strongest. Green practices were directly and significantly related to perceived quality, predominantly indirectly related to satisfaction, and mostly directly related to loyalty. For female users, the relevance of green practices as an antecedent of perceived quality and satisfaction was significantly higher. The results obtained suggest that green practices are an added value that improves users’ perceptions, especially among females, justifying the use of marketing strategies based on the adoption of green practices.

Data Management Analysis for Predicting Stroke using RapidMiner

Stroke are known as the second most leading cause of death. Because of this, data mining techniques are already being used to predict patients that may have stroke. Therefore, we are doing a study to try using data mining techniques using RapidMiner to find information or patterns regarding stroke from a dataset obtained from Kaggle. Three data mining techniques are used in this study, that is classification using decision trees, association rule using FP-Growth algorithm, and clustering technique using k-Means algorithm. Using RapidMiner, we are able to process the dataset using the operators provided in the application. As the result, we found out that due to an unbalanced data, the decision tree model made were only able to predict 68,75% patients as having stroke. With the association rule technique, we found out that most attributes in the dataset does not really associated with each other. With the clustering technique, we were able to group up patients and found out that most patients that have stroke are averaged in the age of 58, with 31 bmi, and 201 average glucose level.

An Explainable Contrastive-based Dilated Convolutional Network with Transformer for Pediatric Pneumonia Detection

Pediatric pneumonia remains a significant global threat, surpassing all other communicable diseases in child mortality and necessitating rapid diagnosis. UNICEF identifies it as a primary cause of death in children under five. However, traditional image processing methods are time-consuming and struggle to capture distinct features due to low-intensity radiation in chest radiographs. Furthermore, skewed training data complicates the extraction of reliable and relevant features. To this end, we propose a novel EXplainable Contrastive-based Dilated Convolutional Network with Transformer (XCCNet) for pediatric pneumonia detection. XCCNet harnesses the spatial power of dilated convolutions and the global insights of contrastive-based transformers for effective feature refinement. A robust chest X-ray processing module tackles low-intensity radiographs, while adversarial-based data augmentation mitigates data distribution challenges. Furthermore, we actively integrate an explainability approach through feature visualization, directly aligning it with the attention region that pinpoints the presence of pneumonia or normality in radiographs. We comprehensively evaluate the efficacy of XCCNet on four publicly available datasets: Kermany, VinDR-PCXR, NIH-Pediatric, and Trivedi, achieved accuracies of 99.76%, 91.56%, 92.87%, and 97.19%, respectively. Quantitative and qualitative assessments demonstrate the effectiveness of XCCNet capabilities. XCCNet’s superior performance enhances the early diagnosis of pneumonia, potentially saving many children’s lives.

Does intellectual property rights protection help reduce carbon emissions?

Our study examines the relationship between intellectual property rights protection (IPRP) and carbon emissions (CE) using an unbalanced panel data set from 116 countries spanning 2008–2020. The analysis reveals a positive relationship between IPRP and CE. Mechanism tests and nonlinear analysis further elucidate this relationship. The mediation effect results indicated that IPRP indirectly fostered CE growth by inhibiting energy structure’s improvement, demonstrating that IPRP exacerbated the dependence on traditional high-carbon energy sources, hindering the transition to cleaner, more efficient energy systems. The analysis also revealed that IPRP indirectly promoted the increase of CE by promoting technological progress and economic growth. The moderation effect results suggest that political stability could mitigate the positive impact of IPRP on CE. Furthermore, the analysis reveals an inverted U-shaped relationship between IPRP and CE, indicating that while IPRP promotes an increase in CE in the initial stages, its intensification suppresses CE growth beyond a turning point. Hence, our study provides new theoretical and empirical evidence to understand the complex relationship between IPRP and CE. It also offers policy recommendations for effectively integrating global carbon reduction efforts with intellectual property policies.

Board gender equality and ESG performance. Evidence from European banking sector

PurposeThe purpose of this paper is to investigate the relationship between board gender equality and environmental, social and governance (ESG) performance in the European banking sector. The study examines whether and how the presence of women on the board of directors (BoD) influences ESG dimensions.Design/methodology/approachThe authors analyzed a sample of 72 European Union banks for the period 2015–2021 and developed an econometric model applying unbalanced panel data regression with firm fixed effects and controls per year. To test the research hypotheses, the authors considered gender equality in terms of female participation on the BoD and measured ESG dimensions by using the ESG score provided by Refinitiv.FindingsThe findings suggest a significant positive relationship between the number of women on BoD and the ESG performance of European banks only up to a certain threshold of female directors (at least three women). The study also explores how the proportion of women on BoD influences the individual ESG pillars. The results show that the percentage of female directors has a positive and statistically significant impact on the social dimension of the ESG framework.Research limitations/implicationsThe investigation is highly relevant to investors considering ESG issues in their decision-making process. The overall findings support policymakers and regulators on how to improve ESG performance through the design and the application of corporate governance (CG) mechanisms. From a managerial perspective, the study suggests that managers and CEOs should focus their efforts on establishing the right gender combination of directors on bank BoDs.Originality/valueThis paper offers an in-depth examination of the CG practices of banks, and it attempts to bridge the gap in prior literature on the determinants of ESG issues in the European banking industry. To the best of the authors’ knowledge, this study is the first that investigates the relationship between the representation of women on BoDs and the ESG dimensions measured by the Refinitiv Eikon score. The use of critical mass theory adds a fresh perspective to the literature on ESG in Europe since the influence of board gender diversity on ESG performance of the European banks is still unaccounted for. This study addresses this pressing research issue drawing on resource dependence, agency and legitimacy theories.

SBD-Net: Incorporating Multi-Level Features for an Efficient Detection Network of Student Behavior in Smart Classrooms

Detecting student behavior in smart classrooms is a critical area of research in educational technology that significantly enhances teaching quality and student engagement. This paper introduces an innovative approach using advanced computer vision and artificial intelligence technologies to monitor and analyze student behavior in real time. Such monitoring assists educators in adjusting their teaching strategies effectively, thereby optimizing classroom instruction. However, the application of this technology faces substantial challenges, including the variability in student sizes, the diversity of behaviors, and occlusions among students in complex classroom settings. Additionally, the uneven distribution of student behaviors presents a significant hurdle. To overcome these challenges, we propose Student Behavior Detection Network (SBD-Net), a lightweight target detection model enhanced by the Focal Modulation module for robust multi-level feature fusion, which augments feature extraction capabilities. Furthermore, the model incorporates the ESLoss function to address the imbalance in behavior sample detection effectively. The innovation continues with the Dyhead detection head, which integrates three-dimensional attention mechanisms, enhancing behavioral representation without escalating computational demands. This balance achieves both a high detection accuracy and manageable computational complexity. Empirical results from our bespoke student behavior dataset, Student Classroom Behavior (SCBehavior), demonstrate that SBD-Net achieves a mean Average Precision (mAP) of 0.824 with a low computational complexity of just 9.8 G. These figures represent a 4.3% improvement in accuracy and a 3.8% increase in recall compared to the baseline model. These advancements underscore the capability of SBD-Net to handle the skewed distribution of student behaviors and to perform high-precision detection in dynamically challenging classroom environments.

DUDS: Diversity-aware unbiased device selection for federated learning on Non-IID and unbalanced data

Federated Learning (FL) is a distributed machine learning approach that preserves privacy by allowing numerous devices to collaboratively train a global model without sharing raw data. However, the frequent exchange of model updates between numerous devices and the central server, and some model updates are similar and redundant, resulting in a waste of communication and computation. Selecting a subset of all devices for FL training can mitigate this issue. Nevertheless, most existing device selection methods are biased, while unbiased methods often perform unstable on Non-Independent Identically Distributed (Non-IID) and unbalanced data. To address this, we propose a stable Diversity-aware Unbiased Device Selection (DUDS) method for FL on Non-IID and unbalanced data. DUDS diversifies the participation probabilities for device sampling in each FL training round, mitigating the randomness of the individual device selection process. By using a leader-based cluster adjustment mechanism to meet unbiased selection constraints, DUDS achieves stable convergence and results close to the optimal, as if all devices participated. Extensive experiments demonstrate the effectiveness of DUDS on Non-IID and unbalanced data scenarios in FL.

Factor Analysis of Ordinal Items: Old Questions, Modern Solutions?

Factor analysis, a staple of correlational psychology, faces challenges with ordinal variables like Likert scales. The validity of traditional methods, particularly maximum likelihood (ML), is debated. Newer approaches, like using polychoric correlation matrices with weighted least squares estimators (WLS), offer solutions. This paper compares maximum likelihood estimation (MLE) with WLS for ordinal variables. While WLS on polychoric correlations generally outperforms MLE on Pearson correlations, especially with nonbell-shaped distributions, it may yield artefactual estimates with severely skewed data. MLE tends to underestimate true loadings, while WLS may overestimate them. Simulations and case studies highlight the importance of item psychometric distributions. Despite advancements, MLE remains robust, underscoring the complexity of analyzing ordinal data in factor analysis. There is no one-size-fits-all approach, emphasizing the need for distributional analyses and careful consideration of data characteristics.

Does Green Finance Development Enhance the Sustainability Performance of China’s Energy Companies?

The achievement of China’s “dual-carbon” standard has been devoted to the green transformation and the sustainable growth of energy firms, both of which can be financed by the growth of green financing. This study aims to investigate how the development level of green finance influences the sustainable development performance of listed energy companies in China. It seeks to delve into the underlying mechanisms connecting green finance with financing constraints and, subsequently, with sustainability performance, as well as exploring the relationship between green finance and green total factor productivity in relation to sustainability performance. Additionally, this study will provide strategies and recommendations to enhance the sustainable development capabilities of energy enterprises. This study empirically evaluates the four aspects of sustainable development performance: economic, social, environmental, and innovative performance—as well as its mechanism of action using the fixed-effects pattern with two ways and the mediated-effects pattern using unbalanced panel data from Chinese-listed energy firms spanning from 2011 to 2020. The study discovered that (1) energy firms’ performance in sustainable development is greatly enhanced by the progression of green finance; (2) the advancement of green finance effectively boosts the sustainable development performance of energy companies by reducing financing constraints and enhancing green total factor productivity; (3) a more distinct relationship is evident between the extent of green financing development and the performance of sustainable development within state-owned enterprises. While green finance development has a stronger role in innovative performance for larger energy firms, it has a noticeable proactive impact on the economic, social, and environmental performance of smaller energy enterprises. Based on the study’s findings, this paper presents recommendations for the enhancement of green financing policies and the sustainable enhancement of energy enterprises in China.

Grow with the flow: Is phenotypic plasticity across hydrodynamic gradients common in seaweeds?

Seaweeds are widely assumed to be phenotypically plastic across hydrodynamic gradients, yet while many marine macroalgae exhibit intraspecific phenotypic variation that correlates with flow, researchers often fail to test whether such variation is due to plasticity or another mechanism, such as local adaptation. In this minireview, we considered mechanisms for sensing flow in seaweeds that could facilitate adaptive phenotypic plasticity across hydrodynamic gradients. We then reviewed the literature from 1900 to 2024 to see how often phenotypic variation and plasticity across hydrodynamic gradients had been observed and demonstrated in different groups of seaweeds. In the last 124 years, phenotypic variation and plasticity in response to flow have been well documented in brown algae but scarcely documented in red and green algae. This could suggest that brown algae are better able to sense and respond to flow than red and green algae, perhaps due to the intercalary meristem of many brown algae, including most kelps. However, this skewed distribution could also be the result of publication bias, as most studies involving flow have been conducted on brown algae. Only 30% of 141 papers specifically investigated if observations of phenotypic variation along hydrodynamic gradients were due to plasticity. To date, phenotypic plasticity in response to flow has been demonstrated in 20 brown algal species, five red algal species, and two green algal species. Thus, the assumption that phenotypic plasticity to flow is common across seaweeds is not particularly well supported by the literature. Mechanisms underlying plasticity to flow are poorly understood and remain a critical avenue for future research.

Reduced copper uptake and efflux by the mussel Mytilus coruscus after Cu exposure: Implication for biomonitoring

The hard-shell mussels Mytilus coruscus have been extensively employed in pollution biomonitoring. Earlier studies indicated that metal concentrations in Mytilus coruscus may not accurately reflect the true metal contamination levels in the sampling areas, possibly due to their modified metal uptake and efflux. Given the likelihood of mussels in the field being exposed to intermittent metal contaminants, this study investigated whether different Cu pre-exposures significantly affected its uptake and efflux upon Cu exposure. We found significant reduction in Cu uptake rate constant (ku) and efflux rate constant (ke) in the mussels with varying Cu pre-exposure regimes. Specifically, the ku decreased from 1.55 ± 0.37 L g−1 d−1 in the control group to 0.65 ± 0.19 after 5 days and 0.53 ± 0.28 after 15 days of exposure to 20 μg L−1 Cu, respectively, and then was further reduced to as low as 0.096 ± 0.046 L g−1 d−1 following a 5-day exposure at 50 μg L−1 Cu. Similarly, the ke decreased from 0.18 ± 0.020 to 0.15 ± 0.015 d−1 following 5–15 days of exposure to 20 μg L−1 Cu, and further decreased to 0.081 ± 0.023 d−1 after a 5-day exposure at 50 μg L−1 Cu. Our subcellular distribution analysis underscored the critical role of the metallothionein-like protein (MTLP) fraction in modifying both Cu ku and ke during the rapid-depuration phase (ke1), whereas the metal-rich granule (MRG) fraction influenced the ke during the second depuration phase (ke2). This study demonstrated that environmental assessments utilizing biomonitoring species should consider the exposure of these organisms to ensure accurate interpretations of metal contamination in marine ecosystems and enhance the effectiveness of these species in environmental monitoring. This crucial factor is often overlooked, potentially skewing data and leading to misinterpretations of environmental health and pollution levels.

Tensor-based unsupervised feature selection for error-robust handling of unbalanced incomplete multi-view data

Recent advancements in multi-view unsupervised feature selection (MUFS) have been notable, yet two primary challenges persist. First, real-world datasets frequently consist of unbalanced incomplete multi-view data, a scenario not adequately addressed by current MUFS methodologies. Second, the inherent complexity and heterogeneity of multi-view data often introduce significant noise, an aspect largely neglected by existing approaches, compromising their noise robustness. To tackle these issues, this paper introduces a Tensor-Based Error Robust Unbalanced Incomplete Multi-view Unsupervised Feature Selection (TERUIMUFS) strategy. The proposed MUFS framework specifically caters to unbalanced incomplete multi-view data, incorporating self-representation learning with a tensor low-rank constraint and sample diversity learning. This approach not only mitigates errors in the self-representation process but also corrects errors in the self-representation tensor, significantly enhancing the model’s resilience to noise. Furthermore, graph learning serves as a pivotal link between MUFS and self-representation learning. An innovative iterative optimization algorithm is developed for TERUIMUFS, complete with a thorough analysis of its convergence and computational complexity. Experimental results demonstrate TERUIMUFS’s effectiveness and competitiveness in addressing unbalanced incomplete multi-view unsupervised feature selection (UIMUFS), marking a significant advancement in the field.