Search

A robust and simple implementation of the generalized Einstein formulation using single equilibrium molecular dynamics simulation is introduced to compute diffusion and shear viscosity. The unique features underlying this framework are as follows: (1) The use of a simple binary-based method to sample time-dependent transport coefficients results in a uniform distribution of data on a logarithmic time scale. Although we sample "on-the-fly," the algorithm is readily applicable for post-processing analysis. Overlapping same-length segments are not sampled as they indicate strong correlations. (2) Transport coefficients are estimated using a power law fitting function, a generalization of the standard linear relation, that accurately describes the long-time plateau. (3) The use of a generalized least squares (GLS) fitting estimator to explicitly consider correlations between fitted data points results in a reliable estimate of the statistical uncertainties in a single run. (4) The covariance matrix for the GLS method is estimated analytically using the Wiener process statistics and computed variances. (5) We provide a Python script to perform the fits and automate the procedure to determine the optimal fitting domain. The framework is applied to two fluids, binary hard sphere and a Lennard-Jones near the triple point, and the validity of the single-run estimates is verified against multiple independent runs. The approach should be applicable to other transport coefficients since the diffusive limit is universal to all of them. Given its rigor and simplicity, this methodology can be readily incorporated into standard molecular dynamics packages using on-the-fly or post-processing analysis. [ABSTRACT FROM AUTHOR]

Purpose: To evaluate the peri-implant parameters of immediately placed and loaded mandibular overdentures over a 5-year follow-up period. Materials and Methods: All subjects who had been advised and planned for two-implant mandibular overdenture treatment were included in this study. The peri-implant parameters –including plaque index (PI), bleeding index (BI) and peri-implant pocket depth (PIPD) as well as marginal bone loss (MBL) – were assessed. In addition, prosthodontic parameters including abutment-, implant- and denture-related complications were assessed. Patients were evaluated at follow-up visits, scheduled at 1, 12, 24, 36, 48, and 60 months. The data distribution was analysed with the Shapiro-Wilk test. Data within follow-up categories were compared using ANOVA and the Tukey-Kramer test. A p-value < 0.05 was considered statistically significant. Results: Among the 32 participants, 19 were males and 13 were females, with a mean age of 60.5 ± 7.33. The mean plaque index (PI), bleeding index (BI) and peri-implant pocket depth (PIPD) varied over time. However, no statistically significant difference was observed in the plaque index, bleeding index and peri-implant pocket depth over time (p > 0.05). The mean value at baseline was found to be -0.9 ± 0.3. The values increased over time, with the highest value observed at 60 months 2.6 ± 0.7, which was statistically significant (p < 0.001). Conclusion: Immediately placed and loaded mandibular implant overdentures using two un-splinted implants with locator attachments showed acceptable PI, BI and PIPD at the 5-year follow-up. Statistically significantly greater marginal bone loss was observed from baseline to follow-up, but it was within acceptable limits. A moderate number of restorative and abutment complications were observed during the follow-up of IODs. [ABSTRACT FROM AUTHOR]

ABSTRACT This article presents a novel approach for fault diagnosis in rotating machinery using the few‐shot learning‐based unknown recognition and classification (FSLB‐UR&C) model. The core contribution of this research lies in addressing the challenges of domain drift, unknown fault detection, and uneven sampling intervals. The model effectively incorporates data scaling using Min–Max scaling to manage vibration data drift without altering the source data distribution, significantly extending the classification range. Principal component analysis visualizations substantiate the superior performance of Min–Max scaling over standard methods in processing drifted data, thus enhancing model accuracy. We validated the model on an in‐house bearing simulator and a coal conveyor motor, demonstrating its enhanced ability to classify drifted data and identify previously unknown faults. Additionally, time interpolation was applied to handle irregular sampling intervals, further optimizing the framework's robustness. Compared to other deep learning techniques, the FSLB‐UR&C model with data scaling exhibits markedly improved performance. [ABSTRACT FROM AUTHOR]

In dynamic ensemble selection (DES) techniques, the competence level of each classifier is estimated from a pool of classifiers, and only the most competent ones are selected to classify a specific test sample and predict its class labels. A significant challenge in DES is efficiently estimating classifier competence for accurate prediction, especially when these techniques employ the K-Nearest Neighbors (KNN) algorithm to define the competence region of a test sample based on a validation set (known as the dynamic selection dataset or DSEL). This challenge is exacerbated when the DSEL does not accurately reflect the original data distribution or contains noisy data. Such conditions can reduce the precision of the system, induce unexpected behaviors, and compromise stability. To address these issues, this paper introduces the self-generating prototype ensemble selection (SGP.DES) framework, which combines meta-learning with prototype selection. The proposed meta-classifier of SGP.DES supports multiple classification algorithms and utilizes meta-features from prototypes derived from the original training set, enhancing the selection of the best classifiers for a test sample. The method improves the efficiency of KNN in defining competence regions by generating a reduced and noise-free DSEL set that preserves the original data distribution. Furthermore, the SGP.DES framework facilitates tailored optimization for specific classification challenges through the use of hyperparameters that control prototype selection and the meta-classifier operation mode to select the most appropriate classification algorithm for dynamic selection. Empirical evaluations of twenty-four classification problems have demonstrated that SGP.DES outperforms state-of-the-art DES methods as well as traditional single-model and ensemble methods in terms of accuracy, confirming its effectiveness across a wide range of classification contexts. [ABSTRACT FROM AUTHOR]

Herb recommendation plays a crucial role in Traditional Chinese Medicine (TCM) by prescribing therapeutic herbs given symptoms. Current herb recommendation methods make promising progress by utilizing graph neural network, but most fail to capture the underlying distribution of the prescription, particularly the long-tailed distribution, and then suffer from cold-start and data sparsity problems such as emerging epidemics or rare diseases. To effectively alleviate these problems, we first propose a novel multi-view contrastive learning method to improve the prediction performance as contrastive learning can derive self-supervision signals from raw data. For exploiting structural and semantic relationship of symptoms and herbs, we construct the symptom-herb graph from inter-view, and the symptom-symptom interactions and the herb-herb interactions from intra-view. From inter-view, we present a new dual structural contrastive learning that adds and drops data depending on the frequency of the prescription dataset to exploit unbalanced data distribution rather than traditional data augmentation. From intra-view, we propose multi-level semantic contrastive learning depending on the co-occurrence frequencies of symptoms and herbs respectively for utilizing various correlations based on statistical results. Experiments conducted on real-world datasets demonstrate the superiority of the proposed method, which improves performance, and robustness against data sparsity. [ABSTRACT FROM AUTHOR]

This article develops a novel estimation framework for the multithreshold accelerated failure time model, which has distinct linear forms within different subdomains. One major challenge is to determine the number of threshold effects. We first show the selection consistency of a modified Bayesian information criterion under mild conditions. It is useful sometimes but heavily depends on the penalization magnitude, which usually varies from the model configuration and data distribution. To address this issue, we leverage a cross‐validation criterion alongside an order‐preserved sample‐splitting scheme to yield a consistent estimation. The new criterion is completely data driven without additional parameters and thus robust to model setting and data distributions. The asymptotic properties for the parameter estimates are also carefully established. Additionally, we propose an efficient score‐type test to examine the existence of threshold effects. The new statistic is free of estimating any potential threshold effects and is thus suitable for multithreshold scenarios. Numerical experiments validate the reliable finite‐sample performance of our methodologies, which corroborates the theoretical results. [ABSTRACT FROM AUTHOR]

ABSTRACT In vehicular ad hoc networks (VANETs), the challenge of broadcast storms during data transmission arises due to an exponential increase in message rebroadcasts. This problem is exacerbated by high‐speed node movements, frequent topology changes, and repetitive discontinuities within these networks, hindering the development of efficient broadcasting protocols. Addressing this gap, our study introduces a pioneering approach utilizing a novel fuzzy method based on multicriteria decision‐making (MCDM) to prioritize vehicles in selecting optimal neighbors for data broadcast. The aim of this work is to propose for VANETs a fuzzy MCDM‐based re‐broadcasting scheme (FMRBS). This method seeks to eliminate broadcast storms and raise data distribution efficiency. We choose the best vehicles for data transportation by using fuzzy logic. The FMRBS system excelled in many respects over UMB and 802.11‐Distance. It decreased end‐to‐end latency and overhead while increasing packet delivery ratio (PDR) and network performance. By efficiently optimizing data distribution inside VANETs, FMRBS lowers broadcasting traffic and network congestion. [ABSTRACT FROM AUTHOR]

Human activity recognition (HAR) using smartphone inertial sensors, like accelerometers and gyroscopes, enhances smartphones' adaptability and user experience. Data distribution from these sensors is affected by several factors including sensor hardware, software, device placement, user demographics, terrain, and more. Most datasets focus on providing variability in user and (sometimes) device placement, limiting domain adaptation and generalization studies. Consequently, models trained on one dataset often perform poorly on others. Despite many publicly available HAR datasets, cross-dataset generalization remains challenging due to data format incompatibilities, such as differences in measurement units, sampling rates, and label encoding. Hence, we introduce the DAGHAR benchmark, a curated collection of datasets for domain adaptation and generalization studies in smartphone-based HAR. We standardized six datasets in terms of accelerometer units, sampling rate, gravity component, activity labels, user partitioning, and time window size, removing trivial biases while preserving intrinsic differences. This enables controlled evaluation of model generalization capabilities. Additionally, we provide baseline performance metrics from state-of-the-art machine learning models, crucial for comprehensive evaluations of generalization in HAR tasks. [ABSTRACT FROM AUTHOR]

Background: Scuba diving was previously excluded because of hypoglycemic risks for patients with type 1 diabetes mellitus(T1DM). Specific eligibility criteria and a safety protocol have been defined, whereas continuous glucose monitoring (CGM) systems have enhanced diabetes management. This study aims to assess the feasibility and accuracy of CGM Dexcom G7 and Free Style Libre 3 in a setting of repetitive scuba diving in T1DM, exploring the possibility of nonadjunctive use. Material and Methods: The study was conducted during an event of Diabete Sommerso® association in 2023. Participants followed a safety protocol, with capillary glucose as reference standard (Beurer GL50Evo). Sensors' accuracy was evaluated through median and mean absolute relative difference (MeARD, MARD) and surveillance error grid (SEG). Data distribution and correlation were estimated by Spearman test and Bland–Altman plots. The ability of sensors to identify hypoglycemia was assessed by contingency tables. Results: Data from 202 dives of 13 patients were collected. The overall MARD was 31% (Dexcom G7) and 14.2% (Free Style Libre 3) and MeARD was 19.7% and 11.6%, respectively. Free Style Libre 3 exhibited better accuracy in normoglycemic and hyperglycemic ranges. SEG analysis showed 82.1% (Dexcom G7) and 97.4% (Free Style Libre 3) data on no-risk zone. Free Style Libre 3 better performed on hypoglycemia identification (diagnostic odds ratio of 254.10 vs. 58.95). Neither of the sensors reached the MARD for nonadjunctive use. Conclusions: The study reveals Free Style Libre 3 superior accuracy compared with Dexcom G7 in a setting of repetitive scuba diving in T1DM, except for hypoglycemic range. Both sensors fail to achieve accuracy for nonadjunctive use. Capillary tests remain crucial for safe dive planning, and sensor data should be interpreted cautiously. We suggest exploring additional factors potentially influencing sensor performance. [ABSTRACT FROM AUTHOR]

Accurate disease recognition through computer vision is crucial for the intelligent management of potato production. Popular data-driven classification methods face challenges including limited labelled data and poor model portability. Unsupervised Domain Adaptation (UDA) addresses these challenges with a novel learning strategy. However, the complex field environment introduces a significant domain shift problem due to varying conditions. Existing UDA methods usually concentrate on aligning global data distribution and employ a single structure for disease feature extraction, thereby limiting their efficacy in true field environment. To tackle this challenge of potato disease recognition, the Multi-Representation Adaptive Network (MRSAN) based on subdomain alignment is presented. MRSAN effectively aligns feature distributions across diverse data by minimising distribution differences among relevant subdomains. Simultaneously, the multi-representation extraction method captures finer details from various perspectives in the disease images. The combination of these two approaches efficiently mitigates the adverse effects caused by various interference factors in field environment. Based on the acquisition conditions of light variation and disease progression, two field potato disease image datasets are created, containing five and six kinds of potato leaf disease, respectively. Extensive transfer experiments are conducted on the two datasets. MRSAN achieves average classification accuracies of 87.03% and 80.06% on the datasets for the corresponding transfer tasks, outperforming the other compared methods. This not only validates the effectiveness of MRSAN but also demonstrates its robust ability to generalise across changes in regard to light variation and disease progression. • A novel method MRSAN based on UDA recognises potato diseases in field environment. • The source domain and target domain are aligned by subdomain alignment. • Lighting changes and disease progression is considered when creating domains. • Multiple feature representations for diseases. • Accuracy surpasses standard pre-training-fine-tuning methods with limited training. [ABSTRACT FROM AUTHOR]

To handle the data imbalance and inaccurate prediction in power load forecasting, an integrated data denoising power load forecasting method is designed. This method divides data into administrative regions, industries, and load characteristics using a four-step method, extracts periodic features using Fourier transform, and uses Kmeans++ for clustering processing. On this basis, a Transformer model based on an adversarial adaptive mechanism is designed, which aligns the data distribution of the source domain and target domain through a domain discriminator and feature extractor, thereby reducing the impact of domain offset on prediction accuracy. The mean square error of the Fourier transform clustering method used in this study was 0.154, which was lower than other methods and had a better data denoising effect. In load forecasting, the mean square errors of the model in predicting long-term load, short-term load, and real-time load were 0.026, 0.107, and 0.107, respectively, all lower than the values of other comparative models. Therefore, the load forecasting model designed for research has accuracy and stability, and it can provide a foundation for the precise control of urban power systems. The contributions of this study include improving the accuracy and stability of the load forecasting model, which provides the basis for the precise control of urban power systems. The model tracks periodicity, short-term load stochasticity, and high-frequency fluctuations in long-term loads well, and possesses high accuracy in short-term, long-term, and real-time load forecasting. [ABSTRACT FROM AUTHOR]

Remote sensing image change captioning (RSICC) has received considerable research interest due to its ability of automatically providing meaningful sentences describing the changes in remote sensing (RS) images. Existing RSICC methods mainly utilize pre-trained networks on natural image datasets to extract feature representations. This degrades performance since aerial images possess distinctive characteristics compared to natural images. In addition, it is challenging to capture the data distribution and perceive contextual information between samples, resulting in limited robustness and generalization of the feature representations. Furthermore, their focus on inherent most change-aware discriminative information is insufficient by directly aggregating all features. To deal with these problems, a novel framework entitled Multi-Attentive network with Diffusion model for RSICC (MADiffCC) is proposed in this work. Specifically, we introduce a diffusion feature extractor based on RS image dataset pre-trained diffusion model to capture the multi-level and multi-time-step feature representations of bitemporal RS images. The diffusion model is able to learn the training data distribution and contextual information of RS objects from which more robust and generalized representations could be extracted for the downstream application of change captioning. Furthermore, a time-channel-spatial attention (TCSA) mechanism based difference encoder is designed to utilize the extracted diffusion features to obtain the discriminative information. A gated multi-head cross-attention (GMCA)-guided change captioning decoder is then proposed to select and fuse crucial hierarchical features for more precise change description generation. Experimental results on the publicly available LEVIR-CC, LEVIRCCD, and DUBAI-CC datasets verify that the developed approach could realize state-of-the-art (SOTA) performance. [ABSTRACT FROM AUTHOR]

Objective: Studies looking into patient and institutional variables linked to extended hospital stays have arisen as a result of the increased focus on severe maternal morbidity and mortality. Understanding the length of hospitalization of patients after delivery is important to gain insights into when hospitals will reach capacity and to predict corresponding staffing or equipment requirements. In Sudan, the distribution of the length of stay during delivery hospitalizations is heavily skewed, with the average length of stay of 2 to 3 days. This study aimed to investigate the use of non-parametric random effect model with Gamma distributed response for analyzing skewed hospital length of stay data in Sudan in neonatal and maternal unit. Methods: We applied Gamma regression models with unknown random effects, estimated using the non-parametric maximum likelihood (NPML) technique [5]. The NPML reduces the heterogeneity in the distribution of the response and produce a robust estimation since it does not require any assumptions on the distribution. The same applies to the log–Gamma link that does not require any transformation for the data distribution and it can handle the outliers in the data points. In this study, the models are fitted with and without covariates and compared using AIC and BIC values. Results: The findings imply that in the context of health care database investigations, Gamma regression models with non–parametric random effect consistently reduce heterogeneity and improve model accuracy. The generalized linear model with covariates and random effect (k = 4) had the best fit, indicating that Sudanese hospital length of stay data could be classified into four groups with varying average stays influenced by maternal, neonatal, and obstetrics data. Conclusion: Identifying factors contributing to longer stays allows hospitals to implement strategies for improvement. Non-parametric random effect model with Gamma distributed response effectively accounts for unobserved heterogeneity and individual-level variability, leading to more accurate inferences and improved patient care. Including random effects can significantly affect variable significance in statistical models, emphasizing the need to consider unobserved heterogeneity when analyzing data containing potential individual-level variability. The findings emphasise the importance of making robust methodological choices in healthcare research in order to inform accurate policy decisions. [ABSTRACT FROM AUTHOR]

In this paper, we propose a generalizable mixed-precision quantization (GMPQ) method for efficient inference. Conventional methods require the consistency of datasets for bitwidth search and model deployment to guarantee the policy optimality, leading to heavy search cost on challenging large-scale datasets in realistic applications. On the contrary, our GMPQ searches the mixed-quantization policy that can be generalized to large-scale datasets with only a small amount of data, so that the search cost is significantly reduced without performance degradation. Specifically, we observe that locating network attribution correctly is general ability for accurate visual analysis across different data distribution. Therefore, despite of pursuing higher accuracy and lower model complexity, we preserve attribution rank consistency between the quantized models and their full-precision counterparts via capacity-aware attribution imitation for generalizable mixed-precision quantization strategy search, where the capacity of quantized networks is considered to fully utilize the network capacity without insufficiency. Since slight noise in attribution is amplified by discrete ranking operations with significant rank errors, mimicking the attribution ranks of the full-precision models obstructs the quantized networks to correctly locate the attribution. To address this, we further present a robust generalizable mixed-precision quantization method to smooth the attribution for rank error alleviation by hierarchical attribution partitioning, which efficiently partitions the attribution pixels in high spatial resolution and assigns the same attribution value for pixels within a group. Moreover, we propose dynamic capacity-aware attribution imitation to adjust the concentration degree of the attribution according to sample hardness, so that sufficient model capacity is achieved with full utilization for each image. Extensive experiments on image classification and object detection show that our GMPQ and R-GMPQ obtain competitive accuracy-complexity trade-offs with significantly reduced search cost compared to the state-of-the-art mixed-precision networks. [ABSTRACT FROM AUTHOR]

Detecting fraud accurately in credit cards is critical as this financial sector incurs significant losses for cardholders. Nonetheless, most studies adopted standard machine learning and few incremental learning, which are inadequate for addressing credit card challenges, such as rapid data arrival, unlimited data, data sensitivity, and performance decline over time. For this purpose, we propose a chunk-based incremental feature learning approach that optimises the fraud model topology for each new chunk and keeps track of one chunk each time. The model consists of several connected sub-models, where a new sub-model is optimally created for each new chunk. To avoid the network growing indefinitely, we limit the number of sub-models. To this end, we retain the most relevant sub-models to the current chunk's data distribution and re-combine them to create the optimal model. We evaluate our approach using two credit card datasets: the first of medium scale contains 2-day payments in 2013, and the second of considerable scale possesses 6-month payments in 2019. We split these datasets into multiple chunks to learn and test incrementally. We compare our approach with static learning methods trained with different scenarios. Moreover, we vary the number of historical sub-models to check their impact on the predictive performance. [ABSTRACT FROM AUTHOR]

Many imputation methods have been developed over the years and tested mostly under ideal settings. Surprisingly, there is no detailed research on how imputation methods perform when the idealized assumptions about the distribution of data and/or model assumptions are partly not fulfilled. This research looks into the susceptibility of imputation techniques, particularly in relation to outliers, misclassifications, and incorrect model specifications. This is crucial knowledge about how well the methods convince in everyday life because, in reality, conditions are usually not ideal, and model assumptions may not hold. The data may not fit the defined models well. Outliers distort the estimates, and misclassifications reduce the quality of most imputation methods. Several different evaluation measures are discussed, from comparing imputed values with true values or comparing certain statistics, from the performance of classifiers to the variance of estimated parameters. Some well-known imputation methods are compared based on real data and simulations. It turns out that robust conditional imputation methods outperform other methods for real data and simulation settings. [ABSTRACT FROM AUTHOR]

The proliferation of Internet of Things (IoT) applications has heightened the vulnerability of information security, making it susceptible to attacks that may lead to the compromise of sensitive data. Intrusion Detection System (IDS) is deployed in IoT networks for the detection of attacks and to ensure the security of information. In previous works, the IDS datasets suffer from an imbalanced distribution of data about attacks and, the flow of packets in IDS which hinders the ability of deep learning models for potent and coherent classification. With the emergence of graph convolution neural network (GCN), a new sub-field of deep learning models, the structure of graphs can be leveraged to represent the data effectively. IDS datasets typically consist of flow records of data which can naturally be represented as graph structures capturing both edge features and network topology information for classification of attacks. Hence, in this paper, a novel GCN-Ensemble fusion model is proposed for enhanced IoT IDS. There are three stages in this proposed model: (1) Data processing and attribute graph generation, (2) Feature engineering and (3) Classification. The flow attributes of data packets in IDS datasets are represented as the edges and the corresponding varying attacks as nodes of the attribute graph. Here the GCN model is leveraged for feature engineering of the IDS dataset. Further, a novel Ensemble of Convolution Neural Networks is proposed for the classification task. The evaluation of the proposed model encompasses the utilization of four distinct datasets, namely BoT-IoT, ToN-IoT, CIC-IDS2018, and NF UQ NIDS. In the BoT-IoT dataset, the proposed model demonstrates superior performance compared to state-of-art models like Deep learning and Graph neural network (GNN), achieving accuracy improvements of 3.16 and 0.91%, respectively. The observed superior performance of the model in comparison to the baseline models serves to emphasize its potential to augment IoT network security. [ABSTRACT FROM AUTHOR]

The proliferation of edge devices and advancements in Internet of Things technology have created a vast array of distributed data sources, necessitating machine learning models that can operate closer to the point of data generation. Traditional centralized machine learning approaches often struggle with real-time big data applications, such as climate prediction and traffic simulation, owing to high communication costs. In this study, we introduce random node entropy pairing, a novel distributed learning method for artificial neural networks tailored to distributed computing environments. This method reduces communication overhead and addresses data imbalance by enabling nodes to exchange only the weights of their local models with randomly selected peers during each communication round, rather than sharing the entire dataset. Our findings indicate that this approach significantly reduces communication costs while maintaining accuracy, even when learning from non-IID local data. Furthermore, we explored additional learning strategies that enhance system accuracy by leveraging the characteristics of this method. The results demonstrate that random node entropy pairing is an effective and efficient solution for distributed learning in environments where communication costs and data distribution present significant challenges. [ABSTRACT FROM AUTHOR]

Cold temperatures harm tropical crops, but understanding how molecular signals help plants cope could aid in climate change adaptation. This study tested if sodium nitroprusside (SNP), a nitric oxide (NO) donor, could improve potato tolerance to cold. Potato seedlings, treated or non-treated (0.5 mM) with SNP, were exposed to cold stress (0 and ‒2 °C) for 6 h. The study was conducted in a completely randomized design, incorporating three factors in three replications. Results showed that cold stress reduced physiological and biochemical parameters in all seedlings, but less so in those treated with SNP. SNP treatment boosted physio-biochemical activity and increased levels of soluble sugars and enzymatic and non-enzymatic antioxidants. Seedlings treated with SNP and exposed to cold stress had lower levels of H2O2 and malondialdehyde, suggesting that NO may alleviate the harmful effects of cold. The analysis conducted using PCA demonstrated correlations between variables and treatment groups. Notably, the first two principal components (PC1 and PC2) accounted for 77.6% and 78.1% of total variance, respectively, under both 0 and − 2 °C temperatures. Under temperatures below 0 °C, the results of the factor analysis (FA) revealed that PC1 exhibited the highest distribution of data, containing the most prominent variation in Squared cosine values (SCV) values at 0.79. Among the variables, Electrolyte leakage (EL) had the best representation in PC1, with the corresponding maximum SCV values at 0.78 under − 2 °C. This result highlights the potential use of SNP in manipulating cold tolerance in potato plants. [ABSTRACT FROM AUTHOR]

Temporal knowledge graphs (TKGs) are a form of knowledge representation constructed based on the evolution of events at different time points. It provides an additional perspective by extending the temporal dimension for a range of downstream tasks. Given the evolving nature of events, it is essential for TKGs to reason about non-existent or future events. Most of the existing models divide the graph into multiple time snapshots and predict future events by modeling information within and between snapshots. However, since the knowledge graph inherently suffers from missing data and uneven data distribution, this time-based division leads to a drastic reduction in available data within each snapshot, which makes it difficult to learn high-quality representations of entities and relationships. In addition, the contribution of historical information changes over time, distinguishing its importance to the final results when capturing information that evolves over time. In this paper, we introduce CH-TKG (Contrastive Learning and Historical Information Learning for TKG Reasoning) to addresses issues related to data sparseness and the ambiguity of historical information weights. Firstly, we obtain embedding representations of entities and relationships with evolutionary dependencies by R-GCN and GRU. On this foundation, we introduce a novel contrastive learning method to optimize the representation of entities and relationships within individual snapshots of sparse data. Then we utilize self-attention and copy mechanisms to learn the effects of different historical data on the final inference results. We conduct extensive experiments on four datasets, and the experimental results demonstrate the effectiveness of our proposed model with sparse data. [ABSTRACT FROM AUTHOR]

Local Interpretable Model-Agnostic Explanations (LIME) is a widely used explainable artificial intelligence (XAI) technique for tabular data. LIME explains how classifiers or regressors make decisions. However, randomization during synthetic data production makes LIME unstable, leading to distrust in its explanations. The objective of this paper is to present STable LIME (ST-LIME). ST-LIME is an extension of LIME that addresses its instability by using fixed noise to generate consistent synthetic data. Additionally, ST-LIME enhances fidelity by employing the standard deviation to create the original data distribution rather than relying on randomization. For each training data feature, calculate standard deviation values with fixed noise to provide unique synthetic data. The paper evaluates the stability of explanations using the Feature Sequence Stability Index (FSSI) and compares ST-LIME against five state-of-the-art techniques on several popular tabular datasets. On average, ST-LIME outperforms these techniques in stability (100%) and fidelity (98.72%), with a low execution time. [ABSTRACT FROM AUTHOR]

Forest cover in Indonesia reached 95.97 million hectares or 51.2% of the total land area. Fluctuations in forest cover from 2014 to 2022 highlight the need for accurate land mapping to understand factors influencing landscape patterns. Decreases in forest cover increase vulnerability to natural disasters, particularly floods, causing significant economic and environmental losses. This study focuses on strategies for land cover mapping, especially in under-monitored regions, leveraging recent advancements in satellite and drone data for rapid and precise mapping. Deep learning methods like convolutional neural networks (CNN) have shown high accuracy in vegetation cover detection, which can be further improved with new activation functions. This study proposes the new modification activation function, with adding a linear component to enhance CNN accuracy and effectiveness. The proposed demonstrated a significant increase in accuracy, maintaining high values during testing (90.07%) on the Vegetation Density of Peatland Drone dataset. Statistical analysis showed high data distribution with significant performance variation. Despite slower computation times, the proposed excelled in convergence speed and effectiveness compared to other activation. ANOVA and Friedman analyses confirmed significant performance and computation time differences among activation methods. The proposed shows excellent potential in enhancing CNN model accuracy and stability for land cover classification. [ABSTRACT FROM AUTHOR]

Copyright of Acta Scientiarum Naturalium Universitatis Sunyatseni / Zhongshan Daxue Xuebao is the property of Sun-Yat-Sen University and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Financial distress prediction (FDP) has attracted high attention from many financial institutions. Utilizing supervised learning‐based methods in FDP, however, is time consuming and labor intensive. Therefore, in this paper, we exploit active‐pSVM method, which combines potential data distribution information and existing expert experience to solve FDP problem. Moreover, with the increasingly popular textual information, we construct several features on our protocol that are based on the Management Discussion and Analysis (MD&A) text information. Using datasets that are collected in different time windows from the listed Chinese companies, we conducted an extensive experiment and were able to confirm a better efficiency of our active‐pSVM, when compared with some common supervised learning‐based methods. Our study also covers the application of MD&A text information on weakly supervised learning model in FDP. [ABSTRACT FROM AUTHOR]

Sports injuries often arise from improper scheduling of exercise loads, and timely assessment of these loads is essential for minimizing injury risk. This article investigates the validity of using changes in pupillary light reflex (PLR) during dynamic aerobic training as a novel approach to evaluating exercise load. Dynamic aerobic training was conducted on a power bicycle for 15 min. With a 3-minute interval as the demarcation line, PLR measurement was performed before training and heart rate was recorded throughout the process. The Rating of Perceived Exertion (RPE) scale invented by Borg was used for scoring before training and after training. The normal distribution of data was confirmed through the Shapiro-Wilk test. Pearson correlation analysis was used to quantify the correlation between variables. The Receiver Operating Characteristic (ROC) curve and the area under the curve (AUC) analysis were used to determine the indicators of exercise load. The optimal threshold of the indicators was calculated through the Youden index to evaluate sensitivity and specificity. Thirty male second-tier athletes with a mean age of 23.66 ± 2.21 years, a mean height of 175.3 ± 6.5 cm, and a mean weight of 68.99 ± 10.35 kg participated in this study. Based on the RPE scale results, it was confirmed that the 15-minute dynamic aerobic exercise successfully elicited varying levels of perceived exertion among the athletes. The findings of this study indicate significant changes in PLR and heart rate (HR) with increasing exercise duration and external load. There were strong correlations between RPE and maximum constriction velocity (MCV) (|r| = 0.8309, p < 0.001, negative correlation), maximum diameter (INIT) (r = 0.7641, p < 0.001, positive correlation), time to reach 75% recovery (T75) (|r| = 0.7289, p < 0.001, negative correlation), and HR (r = 0.8170, p < 0.001, positive correlation). Additionally, the results suggest that MCV is the most significant potential indicator for detecting internal load, exhibiting high specificity and sensitivity (AUC = 0.8509, p < 0.001). Further analysis using the Youden index identified 5.07 mm/s as the optimal cutoff value for MCV, indicating that when MCV ≤ 5.07 mm/s, the athletes' internal load has reached an "Intense" state. PLR may be a potential indicator for assessing internal load Further investigation could involve developing a non-invasive exercise load detection system based on pupillary variable indicators, providing a valuable new approach for accurately measuring exercise load. [ABSTRACT FROM AUTHOR]

Every second, a huge volume of multi-dimensional data is generated in fields such as Social Networking, Industrial Internet of Things, Stock market and E-commerce applications. Knowledge and pattern extraction are a challenging task in the evolving nature of data stream. Major issues are (i) ‘concept drift’ occurs as a result of pattern changes in the data distribution and (ii) ‘concept evolution’ occurs when a new class evolves in the data stream. These issues degrade the performance of learning models. In this paper, we focus on detection of concept evolution and enhance the performance of classifiers. For this, we propose a new model to identify novel classes, namely, Detection of Novel Classes (DNC). The proposed method adopts long short term memory to continuously observe the streaming data in order to detect emerging classes. The continuous monitoring allows the model to distinguish between existing classes and the novel classes which save time and memory. Also, the proposed method is demonstrated for identifying more than one novel class. The experiments are performed over seven different datasets. The results confirm the efficiency is increased ranging from 6% to 34% by the proposed method in identifying new concepts in the evolving data stream than the existing methods available in the literature. [ABSTRACT FROM AUTHOR]

Background: People with aphasia commonly experience depression and anxiety. The individual therapy program, Aphasia PRevention Intervention and Support in Mental health (Aphasia PRISM) offers low intensity psychotherapeutic interventions using communication supports. Trained stroke clinicians from disciplines other than psychology (e.g., allied health professional, nurse, doctor) can deliver such interventions. The aim of the planned studies is to test the acceptability, feasibility and preliminary effectiveness of Aphasia PRISM for preventing depression and anxiety and supporting wellbeing in people with aphasia after stroke. Method: This protocol describes two feasibility studies of the Aphasia PRISM using parallel, convergent, mixed methods designs. Participants will be adults with aphasia after stroke from a metropolitan healthcare site (Study A, pre-post case series, approximate n = 6) and two regional/rural sites (Study B, randomised controlled trial, n = 30; random allocation to either: treatment arm Aphasia PRISM + usual care; or attention control of secondary stroke prevention information + usual care) in Victoria, Australia. Eligible participants will be offered 7 weekly intervention sessions (via telehealth or in person). Other participant groups include stroke clinicians / site managers and family members. Feasibility will be assessed with treatment acceptability ratings (primary outcome) and secondary outcomes such as recruitment rate and participant clinical response. Participant clinical responses to intervention will be collected pre-intervention (baseline), concurrently (4 week acceptability measure), immediately post (8 weeks), follow-up (3 months) and follow-up at 6 months (Study B only). Descriptive statistics will be used to analyse quantitative data based on the type and distribution of data. Reflexive thematic analysis, an inductive approach will be used to analyse qualitative data. Discussion: Aphasia PRISM offers the potential for trained stroke clinicians to deliver low intensity psychotherapeutic interventions with communication supports to people with aphasia. Studies A and B were registered with the Australian New Zealand Clinical Trials registry under one registration (ACTRN12620000209998) 20th February 2020. [ABSTRACT FROM AUTHOR]

Percutaneous nephrolithotomy (PNL) is the preferred surgical procedure for treating renal stones larger than 2 cm. Patients may experience significant pain in the postoperative period due to factors such as the catheter entry site and tension in the renal capsule. Erector Spinae Plane (ESP) block provides perioperative muscle relaxation and reduces analgesic use. In this study, we investigated the effect of ESP block on postoperative pain scale and quality of recovery in patients undergoing PNL. The study was conducted with a total of 96 patients, with 45 patients receiving the ESP block and 51 patients not receiving the ESP block. Following the surgery, the visual pain scale (VAS) was completed at 0, 2, 6, 12, and 24 h, and the Quality of Recovery-15 (QoR-15) scale along with the willingness scale were completed at the 24-hour mark. The normal distribution of data was evaluated with the Kolmogorov-Smirnov test, histogram, and skewness-kurtosis coefficients. For comparing paired groups, the Student's T-test was used for normally distributed variables, while the Mann-Whitney U test was used for parameters that did not have a normal distribution. The comparison of postoperative pain, as measured by VAS scores between the groups, revealed statistically significantly lower scores in the ESP block group at 0, 2, 6, 12, and 24 h (p < 0.001, p < 0.001, p < 0.001, p = 0.008, p = 0.010, respectively). Our study results indicated that patients who received the ESP block experienced reduced pain levels and had higher QoR-15 scores compared to those who did not receive the ESP block. [ABSTRACT FROM AUTHOR]

Background: Spatiotemporal dengue forecasting using machine learning (ML) can contribute to the development of prevention and control strategies for impending dengue outbreaks. However, training data for dengue incidence may be inflated with frequent zero values because of the rarity of cases, which lowers the prediction accuracy. This study aimed to understand the influence of spatiotemporal resolutions of data on the accuracy of dengue incidence prediction using ML models, to understand how the influence of spatiotemporal resolution differs between quantitative and qualitative predictions of dengue incidence, and to improve the accuracy of dengue incidence prediction with zero-inflated data. Methodology: We predicted dengue incidence at six spatiotemporal resolutions and compared their prediction accuracy. Six ML algorithms were compared: generalized additive models, random forests, conditional inference forest, artificial neural networks, support vector machines and regression, and extreme gradient boosting. Data from 2009 to 2012 were used for training, and data from 2013 were used for model validation with quantitative and qualitative dengue variables. To address the inaccuracy in the quantitative prediction of dengue incidence due to zero-inflated data at fine spatiotemporal scales, we developed a hybrid approach in which the second-stage quantitative prediction is performed only when/where the first-stage qualitative model predicts the occurrence of dengue cases. Principal findings: At higher resolutions, the dengue incidence data were zero-inflated, which was insufficient for quantitative pattern extraction of relationships between dengue incidence and environmental variables by ML. Qualitative models, used as binary variables, eased the effect of data distribution. Our novel hybrid approach of combining qualitative and quantitative predictions demonstrated high potential for predicting zero-inflated or rare phenomena, such as dengue. Significance: Our research contributes valuable insights to the field of spatiotemporal dengue prediction and provides an alternative solution to enhance prediction accuracy in zero-inflated data where hurdle or zero-inflated models cannot be applied. Author summary: In our study, we tackled the complex challenge of predicting dengue fever outbreaks, a crucial task in the field of epidemiology. Dengue prediction is complicated because it relies on the quality of data, which may be affected by the temporal and spatial resolution. We explored different machine learning algorithms across various spatial (village, city and region) and temporal resolutions (weekly and monthly). A key hurdle we encountered was the high frequency of zero values in reported dengue cases, a common issue known as zero-inflated data. This phenomenon makes accurate predictions difficult, especially at finer resolutions. To overcome this obstacle, we first made qualitative predictions about the presence or absence of dengue cases. Then, in scenarios indicating disease presence, we estimated the magnitude of cases quantitatively. This innovative method we designated as hybrid approach and significantly enhanced prediction accuracy in zero-inflated data. This approach can be applied to continuous data where zero-inflated or hurdle models cannot be applied. Our findings have broader implications beyond dengue prediction, shedding light on the challenges of dealing with zero-inflated data in various real-world situations. By improving our understanding of these complexities, our research contributes valuable insights that not only benefit scientists working in epidemiology but also have practical applications in public health strategies ensuring more effective and targeted interventions. [ABSTRACT FROM AUTHOR]

The exponential growth of encrypted network traffic poses significant challenges for detecting malicious activities online. The scale of emerging malicious traffic is significantly smaller than that of normal traffic, and the imbalanced data distribution poses challenges for detection. However, most existing methods rely on single-category features for classification, which struggle to detect covert malicious traffic behaviors. In this paper, we introduce a novel semi-supervised approach to identify malicious traffic by leveraging multimodal traffic characteristics. By integrating the sequence and topological information inherent in the traffic, we achieve a multifaceted representation of encrypted traffic. We design two independent neural networks to learn the corresponding sequence and topological features from the traffic. This dual-feature extraction enhances the model's robustness in detecting anomalies within encrypted traffic. The model is trained using a joint strategy that minimizes both the reconstruction error from the autoencoder and the classification loss, allowing it to effectively utilize limited labeled data alongside a large amount of unlabeled data. A confidence-estimation module enhances the classifier's ability to detect unknown attacks. Finally, our method is evaluated on two benchmark datasets, UNSW-NB15 and CICIDS2017, under various scenarios, including different training set label ratios and the presence of unknown attacks. Our model outperforms other models by 3.49% and 5.69% in F1 score at labeling rates of 1% and 0.1%, respectively. [ABSTRACT FROM AUTHOR]

AbstractIn this paper, we develop a novel efficient and robust nonparametric regression estimator under a framework of a feedforward neural network (FNN). There are several interesting characteristics for the proposed estimator. First, the loss function is built upon an estimated maximum likelihood function, which integrates the information from observed data as well as the information from the data distribution. Consequently, the resulting estimator has desirable optimal properties, such as efficiency. Second, different from the traditional maximum likelihood estimation (MLE), the proposed method avoids the specification of the distribution, making it adaptable to various distributions such as heavy tails and multimodal or heterogeneous distributions. Third, the proposed loss function relies on probabilities rather than direct observations as in least square loss, contributing to the robustness of the proposed estimator. Finally, the proposed loss function involves a nonparametric regression function only. This enables the direct application of the existing packages, simplifying the computational and programming requirements. We establish the large sample property of the proposed estimator in terms of its excess risk and minimax near-optimal rate. The theoretical results demonstrate that the proposed estimator is equivalent to the true MLE where the density function is known in terms of excess risk. Our simulation studies show that the proposed estimator outperforms the existing methods based on prediction accuracy, efficiency and robustness. Particularly, it is comparable to the MLE with the known density and even gets slightly better as the sample size increases. This implies that the adaptive and data-driven loss function from the estimated density may offer an additional avenue for capturing valuable information. We further apply the proposed method to four real data examples, resulting in significantly reduced out-of-sample prediction errors compared to existing methods. [ABSTRACT FROM AUTHOR]

ABSTRACT Deep learning has attracted widespread attention in data modeling and key quality indicator prediction in the chemical industry. However, traditional deep learning networks usually distort the original data distribution due to the superposition effect of multiple layers of nonlinear activation functions. In this case, multivariate statistical learning techniques present an avenue to reveal the intrinsic relationship of the data by combining the linear trends between input and predictor variables. To comprehensively capture data features from multiple perspectives, this study proposes a deep learning‐based data modeling network called the information retention unit (IRU). This network combines intrinsic attributes to partial least squares (PLS) and autoencoder (AE) modalities, thus engendering an adaptive response to the complex linear and nonlinear data features. Furthermore, multiple IRUs can be stacked to construct a deep information retention network (DIRN), which enhances the robust extraction of deep data features. Finally, the effectiveness of the proposed network is validated through its prediction application on a dataset obtained from a real‐world chemical industrial process. This method combines multivariate statistical learning techniques based on deep learning, providing an innovative and practical solution for data analysis and prediction in the chemical industry. [ABSTRACT FROM AUTHOR]

An attempt was made to evaluate the elongation level as a stressor on biodegradable starch films reinforced with nanoclay using a simple linear model. A total of 120 film units were subjected to increasing elongation levels and the exact break time of the failed units was monitored. Nine different attempts were made to fit the data distribution and the lognormal distribution was chosen as the most suitable because it resulted in the lowest values of the regression fit indices −2LL, AICc and BIC. Following the selection of the best fit, it was, generally, observed that an increase in the elongation level resulted in the decreasing exact break time of the films. Among several models, the best fit was provided by the simple linear model. Based on this model, the acceleration factor was estimated, and it was shown that it increased exponentially while increasing the elongation level. Finally, the probability of failure and the hazard rate of the film units as a function of the elongation level were estimated, demonstrating the applicability of this method as a tool for food packaging film failure prediction. [ABSTRACT FROM AUTHOR]

Deep learning models are often evaluated in scenarios where the data distribution is different from those used in the training and validation phases. The discrepancy presents a challenge for accurately predicting the performance of models once deployed on the target distribution. Domain adaptation and generalization are widely recognized as effective strategies for addressing such shifts, thereby ensuring reliable performance. The recent promising results in applying vision transformers in computer vision tasks, coupled with advancements in self-attention mechanisms, have demonstrated their significant potential for robustness and generalization in handling distribution shifts. Motivated by the increased interest from the research community, our paper investigates the deployment of vision transformers in domain adaptation and domain generalization scenarios. For domain adaptation methods, we categorize research into feature-level, instance-level, model-level adaptations, and hybrid approaches, along with other categorizations with respect to diverse strategies for enhancing domain adaptation. Similarly, for domain generalization, we categorize research into multi-domain learning, meta-learning, regularization techniques, and data augmentation strategies. We further classify diverse strategies in research, underscoring the various approaches researchers have taken to address distribution shifts by integrating vision transformers. The inclusion of comprehensive tables summarizing these categories is a distinct feature of our work, offering valuable insights for researchers. These findings highlight the versatility of vision transformers in managing distribution shifts, crucial for real-world applications, especially in critical safety and decision-making scenarios. [ABSTRACT FROM AUTHOR]

Goaf instability poses significant hazards, affecting mine safety and public welfare. This study aims to evaluate the risk of goaf instability to enhance safety measures in mining operations. Thirteen key indicators were identified to construct a comprehensive evaluation index system. By integrating game theory, we combined subjective and objective weights to develop a constant weight model, which was subsequently improved by considering data distribution characteristics to develop a local variable weight model. The variable weight intervals were determined through cumulative frequency analysis of normalized factor indices, and the Monte Carlo method was employed to define weight adjustment parameters. Using the cloud model, we assessed the instability risk of goafs. Our results indicate that the variable weight model provides higher evaluation accuracy compared to the constant weight model, offering clearer and more distinguishable membership degrees for the evaluation outcomes, suggesting its potential for more precise risk assessments in mining operations. [ABSTRACT FROM AUTHOR]

ABSTRACT To address the challenges of extracting features from complex industrial process data, the reliance of numerous fault detection methodologies on presupposed data distribution types, and the limited generalization capacity of fault detection, this manuscript introduces a sophisticated algorithm for industrial process fault detection. This algorithm harnesses the information gain adaptive (IGA) technique for feature selection and a synergistic model decision mechanism. Initially, the process involves the computation of information gain via decision trees, coupled with the determination of the k$$ k $$ value through cross‐validation. This strategy enables the adaptive selection of features, thereby facilitating data dimensionality reduction and effective feature extraction. The subsequent phase introduces a ternary statistical measure monitoring group for the detection of linear faults, while autoencoders and one‐class SVM methodologies are applied for the monitoring of nonlinear faults. The culmination of this approach is the development of an innovative weighted decision mechanism, designed to amalgamate the findings from both linear and nonlinear detection avenues, yielding more dependable detection results. The validation of this algorithm employs datasets from the water chillers process and Tennessee Eastman (TE) process, demonstrating the IGA‐combined model's superior performance over isolated linear or nonlinear detection algorithms in terms of detection accuracy and robustness. Notably, the efficacy of this method is not contingent upon specific assumptions regarding data distribution, rendering it a versatile and efficacious tool for the fault detection in industrial processes. [ABSTRACT FROM AUTHOR]

Introduction: The thickness of ground substrate in shallow mountainous areas is a crucial indicator for substrate investigations and a key factor in evaluating substrate quality and function. Reliable data acquisition methods are essential for effective investigation. Methods: This study utilizes six machine learning algorithms -Gradient Boosting Machine (GB), Random Forest (RF), AdaBoost Regressor (AB), Neural Network (NN), Support Vector Machine (SVM), and k-Nearest Neighbors (kNN) -to predict ground substrate thickness. Grid search optimization was employed to finetune model parameters. The models' performances were evaluated using four metrics: mean squared error (MSE), root mean squared error (RMSE), mean absolute error (MAE), and the coefficient of determination (R2). The optimal parameter combinations for each model were then used to calculate the spatial distribution of ground substrate thickness in the study area. Results: The results indicate that after parameter optimization, all models showed significant reductions in the MSE, RMSE, and MAE, while R2 values increased substantially. Under optimal parameters, the RF model achieved an MSE of 1,589, RMSE of 39.8, MAE of 26.5, and an R2 of 0.63, with a Pearson correlation coefficient of 0.80, outperforming the other models. Therefore, parameter tuning is a necessary step in using machine learning models to predict ground substrate thickness, and the performance of all six models improved significantly after tuning. Overall, ensemble learning models provided better predictive performance than other machine learning models, with the RF model demonstrating the best accuracy and robustness. Discussion: Moreover, further attention is required on the characteristics of sample data and environmental variables in machine learning-based predictions. [ABSTRACT FROM AUTHOR]

Spinal medical image segmentation is critical for diagnosing and treating spinal disorders. However, ambiguity in anatomical boundaries and interfering factors in medical images often cause segmentation errors. Current deep learning models cannot fully capture the intrinsic data properties, leading to unstable feature spaces. To tackle the above problems, we propose Verdiff-Net, a novel diffusion-based segmentation framework designed to improve segmentation accuracy and stability by learning the underlying data distribution. Verdiff-Net integrates a multi-scale fusion module (MSFM) for fine feature extraction and a noise semantic adapter (NSA) to refine segmentation masks. Validated across four multi-modality spinal datasets, Verdiff-Net achieves a high Dice coefficient of 93%, demonstrating its potential for clinical applications in precision spinal surgery. [ABSTRACT FROM AUTHOR]

POI data play an important role in various location-based services, including navigation, positioning, and local search applications. However, as cities rapidly develop, a substantial amount of new POI data are generated daily, often accompanied by issues with the quality of their labels. Therefore, there is an urgent need to implement intelligent inference and enhancement processing for POI data labels. Conventional neural network models primarily target balanced data distribution, but they fail to address the issue of imbalanced distribution of POI data labels in terms of quantity. Furthermore, most neural network classification models implicitly learn the semantic knowledge of different categories from training datasets, neglecting the explicit semantic information offered by natural language labels. Considering the above problems, several negative samples are introduced for each input to a positive class, thereby transforming the multi-classification task into a binary classification problem. Simultaneously, POI data labels are introduced to provide explicit semantic information, and the semantic relationship between POI data labels and their names is determined using cross-coding. Experiments demonstrate that the macro − F1 score for the test dataset, which contains 75 different categories of POI data, reaches 0.84. This result surpasses the performance of traditional methods, highlighting the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

Applying evidence theory to structural reliability analysis under epistemic uncertainty, it is necessary to consider the correlation of evidence variables. Among them, solving the joint basic probability assignment (BPA) of the evidence variables is a crucial link. In this study, a solution method of joint BPA based on neural network copula function is proposed. This method is to automatically construct copula function through neural network, which avoids the process of selecting the optimal copula function. Firstly, the neural network copula function is constructed based on the sample set of evidence variables. Then, the expression for solving the joint BPA using the neural network copula function is derived through vectors. Furthermore, the expression is used to map the marginal BPA of evidence variables to joint BPA, thus realizing the solution of joint BPA. Finally, the effectiveness of this method is verified by three examples. The results show that the neural network copula function describes the data distribution better than the optimal copula function selected by the traditional method. In addition, there is actually an error in solving the reliability intervals using the traditional optimal copula function method, whereas the results of this paper's neural network copula function method are more accurate and better for decision making. [ABSTRACT FROM AUTHOR]

The article discusses disruptions in clinical trials caused by external events like pandemics, warfare, and natural disasters, leading to complications such as site closures, supply chain interruptions, and travel restrictions. It explores the use of randomization tests to address disruptions in clinical trials and highlights the importance of incorporating trial disruptions as intercurrent events in the statistical analysis plan. The article also introduces Targeted Learning (TL) as a method to handle disruptions in clinical trials, emphasizing the need for robust analytic strategies in the statistical analysis plan. TL provides a guide for statistical and causal inference, aligning with the principles set forth in the International Council for Harmonization of Technical Requirements for Pharmaceuticals for Human Use (ICH) E9(R1) addendum. [Extracted from the article]

Evaluation of clustering validity to set up an optimal cluster-data space (CDS) is a vital task in many fields related to data mining. Almost existing clustering validity indexes (CVIs) lack stability due to being too sensitive to noise, especially impulse noise. Here, we (1) propose a new CVI named DzI (Dzung Index) or fRisk2 using analysis of fuzzy-set-based accumulated risk degree (FARD), and (2) present a new algorithm named fRisk2-bA for determining the optimal number of data clusters. It is a method of evaluation of the centroid-based fuzzy clustering validity. In essence, the fRisk2 still focuses on enhancing the data compression in each cluster and expanding the separation between cluster centroids. However, these features are exploited indirectly through FARD. As a result, the proposed method not only can avoid the difficulties of the traditional ones relying on the compression and separation properties directly but also can distill better local and global attributes in the data distribution to estimate the CDS more fully. Along with the proved theory basis, surveys, including the ones based on noisy datasets from measurements, showed the compared advantages of fRisk2 as follows. (1) The accuracy, stability, and convergence of the fRisk2 are outstanding. (2) Its total calculating cost is lower than the other surveyed CVIs. [ABSTRACT FROM AUTHOR]

This paper proposes a novel method capable of both detecting OOD data and generating in-distribution data samples. To achieve this, a VAE model is adopted and augmented with a memory module, providing capacities for identifying OOD data and synthesising new in-distribution samples. The proposed VAE is trained on normal data and the memory stores prototypical patterns of the normal data distribution. At test time, the input is encoded by the VAE encoder; this encoding is used as a query to retrieve related memory items, which are then integrated with the input encoding and passed to the decoder for reconstruction. Normal samples reconstruct well and yield low reconstruction errors, while OOD inputs produce high reconstruction errors as their encodings get replaced by retrieved normal patterns. Prior works use memory modules for OOD detection with autoencoders, but this method leverages a VAE architecture to enable generation abilities. Experiments conducted with CIFAR-10 and MNIST datasets show that the memory-augmented VAE consistently outperforms the baseline, particularly where OOD data resembles normal patterns. This notable improvement is due to the enhanced latent space representation provided by the VAE. Overall, the memory-equipped VAE framework excels in identifying OOD and generating creative examples effectively. [ABSTRACT FROM AUTHOR]

How to improve discriminative feature learning is central in classification. Existing works address this problem by explicitly increasing inter-class separability and intra-class compactness by constructing positive and negative pairs for contrastive learning or posing tighter class separating margins. These methods do not exploit the similarity between different classes as they adhere to independent identical distributions assumption in data. In this paper, we embrace the real-world data distribution setting in that some classes share semantic overlaps due to their similar appearances or concepts. Regarding this hypothesis, we propose a novel regularization to improve discriminative learning. We first calibrate the estimated highest likelihood of one sample based on its semantically neighboring classes, then encourage the overall likelihood predictions to be deterministic by imposing an adaptive exponential penalty. As the gradient of the proposed method is roughly proportional to the uncertainty of the predicted likelihoods, we name it adaptive discriminative regularization (ADR), trained along with a standard cross entropy loss in classification. Extensive experiments demonstrate that it can yield consistent and non-trivial performance improvements in a variety of visual classification tasks (over 10 benchmarks). Furthermore, we find it is robust to long-tailed and noisy label data distribution. Its flexible design enables its compatibility with mainstream classification architectures and losses. [ABSTRACT FROM AUTHOR]

In this work, we present a study aimed at the statistical distribution of characteristic signals of laser-induced plasmas. This work mainly focuses on observing statistical distribution for repetitive measurement of spectra, plasma plume imaging, and sound intensity. These were captured by using various laser irradiances, spanning between 1.72 and 6.25 GW cm-2 for a 266 nm laser. Their distributions were fitted by Gaussian, generalized extreme value (GEV), and Burr distributions, as typical representation models used in LIBS. These were compared using the Kolmogorov-Smirnov (KS) test by its null hypothesis on whether these models are suitable or fail to describe the statistical distribution of the data. The behavior of the data distribution has shown a certain connection to the plasma plume temperature. This was observed for all the used ablation energies. Performances of the statistical models were further compared in the outlier filtering process, where the relative standard deviation of the filtered data was observed. The results presented in this work suggest that an appropriate selection of a statistical model for the data representation can lead to an improvement in the LIBS performance. [ABSTRACT FROM AUTHOR]

This article examines the treatment of statistics and error analysis in introductory physics lab manuals. It compares lab manuals from community colleges and universities and identifies inconsistencies in how these topics are addressed. The article suggests that a probabilistic approach to data analysis could enhance the teaching of measurement and uncertainty in the lab. It also highlights the International Organization for Standardization's guide to uncertainty in measurement as a valuable resource for standardizing lab instruction. The article concludes that there is a need for more consistent and rigorous treatment of statistics and error in introductory physics lab manuals. [Extracted from the article]

Many real-world problems have an uneven distribution of data over different classes. The imbalanced nature of data impacts the performance of classifiers. The higher counts of majority class samples influence the learning abilities of well-known classifiers. Genetic programming (GP) algorithm based on natural evolution also impacts if the data's nature is imbalanced. The fitness function plays a pivotal role and impacts almost each building block of the GP framework. GP with the standard fitness function produces under-fitted and biased classifiers. Therefore, this paper has proposed a new fitness function in GP to classify the imbalanced data. The proposed method is used to classify nine imbalanced problems: ABL-18, ABL-9-18, BAL, YEAST2, YEAST1, ABL-9, ION, WDBC, and SPECT. The imbalanced factor of benchmark problems varies from 99:1 to 59:41. The proposed method's performance is compared with K-Nearest-Neighbourhood (KNN) and the standard fitness function-based GP methods. The GP with newly proposed fitness function gives average AUC values for ABL-18(99:1), ABL-9-18(94:6), BAL(92:8), YEAST2(89:11), YEAST1(84:16), ABL-9(83:17), ION(64:36), WDBC(63:37), and SPECT(59:41) as 0.714, 0.812, 0.975, 0.916, 0.768, 0.654, 0.872, 0.939, and 0.704, respectively, which are higher than KNN and the standard fitness function-based GP methods. The result outcomes prove the superiority of the proposed method. [ABSTRACT FROM AUTHOR]

On the basis of data on the geochemical characteristics ((La/Yb)N, Eu/Eu*, and content of Th) of clay rocks of Podolian Transnistria, Southeastern Poland, Belarus and Volyn, Lithuania, the vicinity of St. Petersburg, Southeastern White Sea region, and Arctic Norway, categories of rivers that transported fine-grained aluminosilicate clastic material to sinks that existed during the Vendian and Early Cambrian were reconstructed in the west of the East European Platform. The distribution of data points of clay rocks of the Volhynian time on the (La/Yb)N–Eu/Eu* diagram with the fields of the composition of the pelitic fraction of bottom sediments in the mouth of modern rivers of various categories allows us to assume that, for Lithuania and Podolian Transnistria, the feeding provinces were, among other things, paleowatersheds composed of volcanic formations. For Eastern Belarus and the Southeastern White Sea region, the influence of erosion products supplied by rivers flowing through igneous/metamorphic terranes (crystalline basement) is noticeable. A significant part of the clastic material was carried by rivers that drained sedimentary formations, as well as large rivers, i.e., rivers whose length exceeded 1000 km and whose drainage area was more than 100 000 km2. The existence of the latter is confirmed by the presence of detrital zircon in the rocks of the Zhukov Formation of Southeastern Poland, borrowed possibly from the rocks of Fennoscandia. In the Redkinian time, along with large rivers (rivers of category 1) and rivers fed by the products of erosion of sedimentary formations (rivers of category 2), transport of fine-grained clastic rocks was also carried out by rivers that drained rocks of the crystalline basement (rivers of category 3; this is typical of Belarus and Volyn, the Southeastern White Sea, and Arctic Norway) and by rivers flowing through areas of distribution of volcanic associations (rivers of category 4). The distribution of data points of the clay rocks of the Kotlin stage on the (La/Yb)N–Eu/Eu* diagram suggests that the main agents of transport of fine-grained aluminosilicate clastic material to sinks at that time were large rivers and rivers fed by fine-grained particulate matter owing to the erosion of predominantly sedimentary rocks. The Early Cambrian paleowatersheds were apparently composed not only of crystalline rocks but also of sedimentary formations. All of the above is in fairly good agreement with the previously established fact of a gradual increase in the contribution of erosion products of sedimentary rocks to the formation of Vendian–Early Cambrian sedimentary sequences from Podolian Transnistria to Arctic Norway. [ABSTRACT FROM AUTHOR]