Search

Estimating ranks, quantiles, and distributions over streaming data is a central task in data analysis and monitoring. Given a stream of n items from a data universe equipped with a total order, the task is to compute a sketch (data structure) of size polylogarithmic in n. Given the sketch and a query item y, one should be able to approximate its rank in the stream, i.e., the number of stream elements smaller than or equal to y. Most works to date focused on additive εn error approximation, culminating in the KLL sketch that achieved optimal asymptotic behavior. This article investigates multiplicative (1±ε )-error approximations to the rank. Practical motivation for multiplicative error stems from demands to understand the tails of distributions, and hence for sketches to be more accurate near extreme values. The most space-efficient algorithms due to prior work store either O(log(ε²n)/ε²) or O(log³ (εn)/ε ) universe items. We present a randomized sketch storing O(log1.5 (εn)/ε ) items that can (1 ± ε )-approximate the rank of each universe item with high constant probability; this space bound is within an O(√log(εn)) factor of optimal. Our algorithm does not require prior knowledge of the stream length and is fully mergeable, rendering it suitable for parallel and distributed computing environments. [ABSTRACT FROM AUTHOR]

The increasing cognitive load on infantry squad leaders is a common challenge in modern military operations. As this can increase health and safety risks, there is a need to study the factors responsible for the increase in cognitive load. Ecological situations inherently lack strong experimental controls; therefore, microworlds that simulate real tasks are the usual alternative to field studies. However, to the best of our knowledge, there are currently no microworlds that reproduce the main tasks of the squad leader during operations. This article adresses this gap by describing the design and validation of a new microworld: the Simulated Multitasking Environment for the Squad leader (SMES). Qualitative research was firstly conducted to highlight several squad leader's generic tasks (i.e., common to many situations in the field) that guided the design of the SMES. Psychometric validation of the SMES was then based on two experiments: (i) the first evaluated the microworld's psychometric qualities when tasks were performed individually; and (ii) the second explored concurrent tasks, reflecting real-world complexity. The results showed that the parameters manipulated for each task were relevant for inducing cognitive load, measured using a secondary detection response task and the NASA-TLX questionnaire. The SMES demonstrated satisfactory convergent and content validity in multitasking but not in single-task conditions. Performance in multitasking situations therefore does not seem to depend on task-specific skills, suggesting the existence of an independent factor-multitasking ability. Theoretical and practical implications of the SMES validation are discussed. [ABSTRACT FROM AUTHOR]

To improve the motivation and enthusiasm of subjects during active rehabilitation training, this paper proposes a novel training task planning-based adaptive assist-as-needed (TTP-AAAN) control algorithm for an upper limb exoskeleton. The overall controller contains an outer control loop to determine the required assistive force, and an inner control loop to drive the exoskeleton to track subject motion and to provide desired assistive force obtained from the outer control loop. In the outer control loop, a motion intention and task performance evaluation (MITPE) strategy is established to learn the motor capability of the subject. Based on the obtained evaluation result, the radius and frequency of multi-periodic trajectory tracking task, and the gain of the assistive force are adaptively adjusted by using the adaptive central pattern generator (ACPG) algorithm. Then, in the inner control loop, an asymmetric barrier Lyapunov function-based adaptive output feedback (ABLF-AOF) controller, in combination with a neural network (NN) state observer, is developed. The exoskeleton tracking errors are constrained by the asymmetric barrier Lyapunov function, and the state variables and uncertainty terms of the exoskeleton are simultaneously estimated by the NN state observer. Experiments are carried out with an upper limb exoskeleton to demonstrate the effectiveness of the proposed control strategy. The experimental results show that the developed control scheme can provide assistance and achieve task parameter adaption for the subjects with different motion patterns. In addition, the proposed controller has better training performance than task performance-based adaptive velocity assist-as-needed (AAN) controller and minimal AAN controller. [ABSTRACT FROM AUTHOR]

Recently, multi-modal vision-language foundation models have gained significant attention in the medical field. While these models offer great opportunities, they still face crucial challenges, such as the requirement for fine-grained knowledge understanding in computer-aided diagnosis and the capability of utilizing very limited or even no task-specific labeled data in real-world clinical applications. In this study, we present MaCo, a masked contrastive chest X-ray foundation model that tackles these challenges. MaCo explores masked contrastive learning to simultaneously achieve fine-grained image understanding and zero-shot learning for a variety of medical imaging tasks. It designs a correlation weighting mechanism to adjust the correlation between masked chest X-ray image patches and their corresponding reports, thereby enhancing the model's representation learning capabilities. To evaluate the performance of MaCo, we conducted extensive experiments using 6 well-known open-source X-ray datasets. The experimental results demonstrate the superiority of MaCo over 10 state-of-the-art approaches across tasks such as classification, segmentation, detection, and phrase grounding. These findings highlight the significant potential of MaCo in advancing a wide range of medical image analysis tasks. Multi-modal foundation models are increasingly important in medical applications. Here, authors show a masked contrastive chest X-ray model that achieves fine-grained image understanding and zero-shot capabilities, outperforming existing methods [ABSTRACT FROM AUTHOR]

Microplastic analysis is a crucial step for locating the environmental contamination sources and controlling plastic contamination. A popular tool like Fourier transform infrared (FTIR) spectroscopy is capable of identifying plastic types and can be carried out through a variety of containers. Unfortunately, sample collection from water sources like rivers usually involves filtration so the measurements inevitably include the membrane filter that also has its own FTIR characteristic bands. Furthermore, when plastic particles are small, the membrane filter's spectrum may overwhelm the desired plastics' spectrum. In this study, we proposed a novel preprocessing method based on the dictionary learning technique for decomposing the variations within the acquired FTIR spectra and capturing the membrane filter's characteristic bands for the effective removal of these unwanted signals. We break down the plastic analysis task into two subtasks — membrane filter removal and plastic classification — to increase the explainability of the method. In the experiments, our method demonstrates a 1.5-fold improvement compared with baseline, and yields comparable results compared to other state-of-the-art methods such as UNet when applied to noisy spectra with low signal-to-noise ratio (SNR), but offers explainability, a crucial quality that is missing in other state-of-the-art methods. The limitations of the method are studied by testing against generated spectra with different levels of noise, with SNR ranging from 0 to – 30dB, as well as samples collected from the lab. The components/atoms learned from the dictionary learning technique are also scrutinized to describe the explainability and demonstrate the effectiveness of our proposed method in practical applications. [ABSTRACT FROM AUTHOR]

Emotion-focused therapy (EFT) is an empirically validated form of psychotherapy that has been indicated in the treatment of eating disorders given the alexithymic nature of individuals with eating disorders, particularly anorexia. There were two aspects to the single case study of a client diagnosed with anorexia nervosa who took part in an emotion-focused therapy across 28 months. First, the Observer-Rated Measure of Affect Regulation (O-MAR) was used to track change in emotion experience across treatment. Second, a task analytic research methodology was employed to track client process as she engaged in the two-chair dialog for self-evaluative splits throughout therapy. Implications for updating the model for working with self-evaluative splits in application to this specific population are outlined. The detailed case study is intended to serve as a stepping stone for future research regarding efficacious processes when using EFT with eating disorder clients, particularly those whose suffering has been chronic and severe. [ABSTRACT FROM AUTHOR]

The lock-in amplifier (LIA) instruments are designed to provide signal conditioning for precision measurement systems to extract signals from extremely noisy environments. The digital LIAs design often requires a verification process to ensure hardware performance. Thus, hardware description language (HDL) with functional verification strategies offers a powerful tool to provide an field-programmable gate array (FPGA) integrated solution. In this letter, we propose a methodology of design and verification of all-digital LIA and an additive white Gaussian noise (AWGN) module able to measure extremely lower levels of signal-to-noise ratio (SNR) of $\approx $ ${10}^{-{15}}$ or down to −37 dB while a wide reserve of spurious-free dynamic range (SFDR) up to 90 dB on FPGA is ensured. To this end, the designed and implemented FPGA framework for quick, accurate, and comprehensive characterization of a given digital LIA is used to leverage the capabilities of the design under controllable AWGN noise patterns stimulus. [ABSTRACT FROM AUTHOR]

Finite-field arithmetic operations are vital for the computation of complex cryptography algorithms used in several cutting-edge applications, such as side-channel attacks, authentication, and digital signatures, among others. Currently, the simulation of these algorithms exceeds the computational capabilities of conventional computing systems. This aspect becomes critical, especially when these algorithms are implemented in resource-constrained electronic appliances. In particular, the improvement of execution time in these devices generally require more area. To overcome this issue, a large number of works have been focused on the development of compact conventional binary finite-field arithmetic circuits over GF(p) since these demand a large area consumption. Inspired by neural phenomena, a new emerging branch of computer science has made intensive efforts to improve area consumption of conventional arithmetic circuits. However, the development of compact finite-field arithmetic circuits over GF(p) is a still a challenging task. In this letter, we present for the first time, the design of four new finite-field arithmetic circuits over GF(p) based on spiking neural P (SN P) systems with communication on request. In addition, we propose a neural processor to perform four new finite-field arithmetic operations over GF(p) by using the same processing core, which is not feasible with the use of conventional binary circuits since each finite-field arithmetic-binary circuit over GF(p) is implemented separately, to significantly improve the area consumption. This has mainly been achieved since the neural processor dynamically change its configuration, which is defined in terms of the connectivity and firing rules of each neuron. [ABSTRACT FROM AUTHOR]

Programming for swarm robots is challenging due to platform diversity and the gap between individual and swarm behaviors. To tackle this challenge, we propose a component-based software synthesis method from a high-level specification. To support heterogeneous robots and maximize code reuse, we adopt a component-based approach that classifies software components into three categories: 1) robot; 2) algorithm; and 3) consensus. We generate a task graph model for an individual robot from a high-level specification and use a software synthesizer to generate the target code from the task graph model. Through a proof-of-concept implementation with a group searching application, the viability of the proposed technique is demonstrated. [ABSTRACT FROM AUTHOR]

Crowdsourcing holds broad applications in information acquisition and dissemination, yet encounters challenges pertaining to data quality assessment and user reputation management. Reputation mechanisms stand as crucial solutions for appraising and updating participant reputation scores, thereby elevating the quality and dependability of crowdsourced data. However, these mechanisms face several challenges in traditional crowdsourcing systems: 1) platform security lacks robust guarantees and may be susceptible to attacks; 2) there exists a potential for large-scale privacy breaches; and 3) incentive mechanisms relying on reputation scores may encounter issues as reputation updates hinge on task demander evaluations, occasionally lacking a dedicated reputation update module. This paper introduces a reputation update scheme tailored for crowdsourcing, with a focus on proficiently overseeing participant reputations and alleviating the impact of malicious activities on the sensing system. Here, the reputation update scheme is determined by an Empirical Cumulative distribution-based Outlier Detection method (ECOD). Our scheme embraces a blockchain-based crowdsourcing framework utilizing a homomorphic encryption method to ensure data transparency and tamper-resistance. Computation of user reputation scores relies on their behavioral history, actively discouraging undesirable conduct. Additionally, we introduce a dynamic weight incentive mechanism that mirrors alterations in participant reputation, enabling the system to allocate incentives based on user behavior and reputation. Our scheme undergoes evaluation on 11 datasets, revealing substantial enhancements in data credibility for crowdsourcing systems and a reduction in the influence of malicious behavior. This research not only presents a practical solution for crowdsourcing reputation management but also offers valuable insights for future research and applications, holding promise for fostering more reliable and high-quality data collection in crowdsourcing across diverse domains. [ABSTRACT FROM AUTHOR]

The flanker task (Eriksen & Eriksen, Perception & Psychophysics, 16(1), 143-149, 1974) has been highly influential and widely used in studies of visual attention. Its simplicity has made it popular to include it in experimental software packages and online platforms. The spacing flanker task (SFT), in which the distance between the target and flankers varies, is useful for studying the distribution of attention across space as well as inhibitory control. Use of the SFT requires that the viewing environment (e.g., stimulus size and viewing distance) be controlled, which is a challenge for online delivery. We implement and evaluate an online version of the SFT that includes two calibration pretests to provide the necessary control. Test–retest and split-half reliability of the online version was compared with a laboratory version on measures of inhibitory control and measures of the distribution of attention across space. Analyses show that the online SFT is comparable to laboratory testing on all measures. Results also identify two measures with good test–retest reliability that hold promise for studying performance in the SFT: the mean flanker effect (ICC = 0.745) and RTs on incongruent trials across distances (ICC = 0.65–0.71). [ABSTRACT FROM AUTHOR]

The specific learning disabilities (SLD) identification literature is replete with competing narratives concerning the advantages and disadvantages of various techniques and methods. Until a widely accepted and empirically proven SLD identification methodology is universally supported, evaluators should seek to improve the existing alternatives. This article describes the value of using norm-referenced testing of intellectual development to comprehensively identify specific learning disabilities (SLD) as advocated by the Core-Selective Evaluation Process (C-SEP). To this end, we will define intellectual development and describe practices such as integrated data analysis and task demand analysis. [ABSTRACT FROM AUTHOR]

Claims analysis and risk management is key to avoiding fraud and managing risk in the life insurance industry. Though visualization plays a fundamental role in supporting analysis tasks in the business domain, exploring user behaviour remains a challenging task. The prevalence of natural language interactions enhanced with data visualization has become quite the norm. With the increasing demand for visualization tools and varying levels of user expertise, the use of a natural language interface (NLI) is common. However, the design of visual analytics tools enhanced by NLIs for risk management and claim analysis requires thorough task analysis and domain expertise. In this work, we investigate an alternative approach through a natural language interaction-based interactive visualization such as a chart, pie chart, or histogram, which can be used for insurance claim analysis and risk management. We design a new visual analytics solution named InsCRMVis which is enhanced by NLIs. We present an expert evaluation of InsCRMVis based on the Task Load Index questionnaire which is currently the standard metrics measure to investigate its usability (effectiveness, efficiency, satisfaction) to support human task performance. To evaluate its performance, we performed a user study of ten experts which suggests that InsCRMVis can provide better insights and assistance to insurance managers in reducing loss and guiding changes to insurance premium policies. Furthermore, we provide a concise set of guidelines that can be used when visualizing risk to avoid dangers in the insurance domain. We discuss the challenges associated with the use of a visualization system in the insurance industry, focusing on aspects related to visualization research. [ABSTRACT FROM AUTHOR]

Building task-oriented dialogue systems has become a topic of interest in the research community and industry. The task-oriented dialogue system is a closed-domain dialogue system that can perform specific tasks for users. The natural language understanding module of a task-oriented dialogue system is crucial because it is related to a task-oriented dialogue system that provides correctional services for users. The natural language understanding module of a task-oriented dialogue system performs two tasks: intent detection and slot filling. The intent detection task can be regarded as a text classification task; a classification model is trained to predict the intention of the user from the user's input information. The slot filling task can be regarded as a sequence analysis task; a sequence analysis model is trained to predict the details of the user's intention. In this paper, we proposed a novel model based on a transformer encoder for intent detection and slot filling. It follows the encoder-decoder structure, including a vanilla Transformer encoder, a bidirectional LSTM encoder, a linear classification decoder for intent detection, and a conditional random field decoder for slot filling. The experimental results on two public datasets show that our proposed model outperforms the existing methods based on the Transformer and can be combined with BERT to achieve better intent detection and slot filling results. [ABSTRACT FROM AUTHOR]

Summary: Hospital readmission prediction is defined as an evaluation task to model the historical medical data to predict whether patients will be readmitted after discharge. In the past few years, many feasible and effective prediction methods have been proposed, however, most of them neglect the imbalanced distribution of medical data, which causes great difficulties in modeling. Thus, we proposed a new hospital readmission prediction method, which utilizes hybrid‐sampling and self‐paced balance learning strategies to solve the class‐imbalance problem. To be specifically, we first employ an interference negative sample deletion strategy to reduce the probability of important majority class samples being deleted. Then, we design a hard positive sample generation strategy to generate more positive samples. Meanwhile, we also introduce a self‐paced balance factor during the oversampling process to improve the similarity between newly generated minority class samples and hard positive samples. Finally, we perform the experiments on six real‐world readmission datasets to indicate the superiority of the proposed method. [ABSTRACT FROM AUTHOR]

The COVID-19 pandemic has profoundly impacted our daily lives, disrupting the usual course of life as we know it. The persistent and unpredictable nature of the virus outbreak, coupled with the challenges in finding effective treatments and a satisfactory cure rate, has given rise to an environment that significantly affects our mental well-being. This article focuses on analyzing sentiments expressed in thousands of tweets related to the COVID-19 pandemic. To accomplish automated tweet sentiment classification, a novel deep learning model called CovDLCNet is introduced. This model is based on the classical Long Short-Term Memory (LSTM) network, which has proven to be effective in sequence analysis tasks. By employing Python-based libraries, the necessary data acquisition, data pre-processing, and data analysis processes are carried out. As a data preprocessing task tweets are initially cleaned and for removing the duplicate (redundant) and similar types of tweets one graph theoretic approach is proposed. Further by calculating the polarity score, sentiment of the tweets are categorized into three groups (viz., negative, neutral and positive). Thereafter, by applying the proposed model classifications of the tweets are evaluated. Experimental results show that the proposed model outperforms all other compared models by achieving a 99.23% classification accuracy score. Henceforth, the results suggest that the proposed model performed better than other six compared models viz., LSTM-M1, LSTM-M2, CNN-LSTM, Bernoulli Naïve Bayes (BNB), Logistic regression (LR) and Random forest (RF) for classifying the sentiment of the tweets into negative, neutral and positive categories. [ABSTRACT FROM AUTHOR]

Entity recognition and extraction from contracts play a crucial role in automating contract analysis and extracting valuable information. Named Entity Recognition (NER) techniques are used for identifying and classifying specific entities such as parties, dates, amounts, and clauses within contracts. In this study, we create a high-quality NER dataset from various types of English language contracts by considering their structure, and the legal terminology used within these documents. We present a systematic approach to manually annotate contracts with appropriate entity labels, ensuring accuracy and consistency. The resulting NER dataset serves as a valuable resource for training and evaluating NER models for contract analysis tasks. We evaluate the performance of NER on this dataset using a range of methods. These methods include Conditional Random Fields, various Bidirectional LSTM configurations, and BERT models. Each of these models brings different strengths and capabilities to the task of entity recognition, allowing for a comprehensive evaluation and the selection of the best models over the dataset. Among these, the NER model based on Contracts–BERT–base from the Legal–BERT family, which is pre-trained specifically on English contracts, outperformed all others, achieving an impressive overall F1 score of 0.94. [ABSTRACT FROM AUTHOR]

An increasingly common viewpoint is that protein dynamics datasets reside in a nonlinear subspace of low conformational energy. Ideal data analysis tools should therefore account for such nonlinear geometry. The Riemannian geometry setting can be suitable for a variety of reasons. First, it comes with a rich mathematical structure to account for a wide range of geometries that can be modeled after an energy landscape. Second, many standard data analysis tools developed for data in Euclidean space can be generalized to Riemannian manifolds. In the context of protein dynamics, a conceptual challenge comes from the lack of guidelines for constructing a smooth Riemannian structure based on an energy landscape. In addition, computational feasibility in computing geodesics and related mappings poses a major challenge. This work considers these challenges. The first part of the paper develops a local approximation technique for computing geodesics and related mappings on Riemannian manifolds in a computationally feasible manner. The second part constructs a smooth manifold and a Riemannian structure that is based on an energy landscape for protein conformations. The resulting Riemannian geometry is tested on several data analysis tasks relevant for protein dynamics data. In particular, the geodesics with given start- and end-points approximately recover corresponding molecular dynamics trajectories for proteins that undergo relatively ordered transitions with medium-sized deformations. The Riemannian protein geometry also gives physically realistic summary statistics and retrieves the underlying dimension even for large-sized deformations within seconds on a laptop. [ABSTRACT FROM AUTHOR]

Introduction: Cognitive models have proposed that behavioral tasks can be categorized along at least three dimensions: the sensory-motor modality of the information, its representational format (e.g., location vs. identity), and the cognitive processes that transform it (e.g., response selection). Moreover, we can quickly and flexibly encode, represent, or manipulate information along any of these dimensions. How is this flexibility in encoding such information implemented in the cerebral cortex? Methods: To address this question, we devised a series of functional magnetic resonance imaging (fMRI) experiments in each of which participants performed two distinct tasks that diered along one of the three dimensions. Results: Using multivariate pattern analysis of the fMRI data, we were able to decode between tasks along at least one task dimension within each of the cortical regions activated by these tasks. Moreover, the multiple demand network, a system of brain regions previously associated with flexible task encoding, was largely composed of closely juxtaposed sets of voxels that were specialized along each of the three tested task dimensions. Discussion: These results suggest that flexible task encoding is primarily achieved by the juxtaposition of specialized representations processing each task dimension in the multiple demand network [ABSTRACT FROM AUTHOR]

Attributed graph clustering or community detection which learns to cluster the nodes of a graph is a challenging task in graph analysis. Recently contrastive learning has shown significant results in various unsupervised graph learning tasks. In spite of the success of graph contrastive learning methods in self-supervised graph learning, using them for graph clustering is not well explored. In this paper, we introduce a contrastive learning framework for learning clustering-friendly node embedding. We propose Gaussian mixture information maximization which utilizes a mutual information maximization approach for node embedding. Meanwhile, in order to have a clustering-friendly embedding space, it imposes a mixture of Gaussians distribution on this space. The parameters of the contrastive node embedding model and the mixture distribution are optimized jointly in a unified framework. Experiments show that our clustering-directed embedding space can enhance clustering performance in comparison with the case where community structure of the graph is ignored during node representation learning. The results on real-world datasets demonstrate the effectiveness of our method in community detection. [ABSTRACT FROM AUTHOR]

Graph clustering is an essential task in data analysis. Recently, there has been a notable trend in the application of deep learning in graph clustering. However, it is worth noting that existing deep graph clustering methods primarily focus on the topological information of nodes while overlooking the use of attribute information in the learned feature representations. In addition, they do not address the consistency of the space distribution in attribute and structure clustering results. To tackle these critical issues, a novel structured deep graph clustering network with consistency constraint (CC-DGC) is proposed. The network first constructs an autoencoder to learn and transmit hierarchical attribute information to the graph autoencoder (GAE). Subsequently, the GAE integrates the received hierarchical attribute information with extracted topological information to generate enhanced clustering representations. Moreover, this paper designs a consistency constraint module to promote consistency between the autoencoder and the GAE by optimizing the cluster space distributions they produce. Finally, the feature extraction and clustering classification processes are synchronized and optimized in a self-supervised manner within a unified framework. Extensive experiments illustrate that the proposed CC-DGC demonstrates superiority over the state-of-the-art deep graph clustering methods on five benchmark datasets. [ABSTRACT FROM AUTHOR]

The Machine Learning domain has made a major process with the progression of state-of-the-art technologies. Since current algorithms often don’t provide palatable learning performance, it is necessary to continually upgrade them. This paper has illustrated the comparison of the Long Short-Term Memory (LSTM) model and the Artificial Neural Networks (ANN) model in the prediction of the Internet Movie Database (IMDb) website. These evaluations were then related to sentiment assessment approaches to evaluate their predicted accuracy and performances. The results demonstrate that the ANN model outperforms the LSTM model with a high accuracy rate in terms of the prediction accuracy and loss indicators for the IMDb movie review’s sentiment analysis task in terms of the prediction accuracy and loss indicators for the IMDb movie review’s sentiment analysis task. The accuracy of prediction on the test dataset of the ANN model is 83.5 % and the LSTM model is 83.5%. Therefore, it can be concluded that the standard artificial neural network model that was utilized is an appropriate technique for sentiment assessment tasks in IMDb rating text data. [ABSTRACT FROM AUTHOR]

Online news platforms have become users' primary information sources. However, they focus on attracting users to click on the news and ignore whether the news triggers a sense of engagement, which could potentially reduce users' participation in public events. Therefore, this study constructs four indicators by assessing user engagement to build an intelligent system to help platforms optimize their publishing strategies. First, this study defines user engagement evaluation as a classification task that divides user engagement into four indicators and proposes an extended LDA model based on user click–comment behavior (UCCB), using which the attractiveness of words in news headlines and content can be effectively represented. Second, this study proposes a deep user engagement evaluation (DUEE) model that integrates news attractiveness and multiple features in an attention-based deep neural network for user engagement evaluation. The DUEE model considers various elements that collectively determine the ability of the news to attract clicks and engagement. Third, the proposed model is compared with the baseline and state-of-the-art techniques, showing that it outperforms all existing methods. This study provides new research contributions and ideas for improving user engagement in online news evaluation. [ABSTRACT FROM AUTHOR]

Computational models of the cardiovascular system are increasingly used for the diagnosis, treatment, and prevention of cardiovascular disease. Before being used for translational applications, the predictive abilities of these models need to be thoroughly demonstrated through verification, validation, and uncertainty quantification. When results depend on multiple uncertain inputs, sensitivity analysis is typically the first step required to separate relevant from unimportant inputs, and is key to determine an initial reduction on the problem dimensionality that will significantly affect the cost of all downstream analysis tasks. For computationally expensive models with numerous uncertain inputs, sample‐based sensitivity analysis may become impractical due to the substantial number of model evaluations it typically necessitates. To overcome this limitation, we consider recently proposed Multifidelity Monte Carlo estimators for Sobol' sensitivity indices, and demonstrate their applicability to an idealized model of the common carotid artery. Variance reduction is achieved combining a small number of three‐dimensional fluid–structure interaction simulations with affordable one‐ and zero‐dimensional reduced‐order models. These multifidelity Monte Carlo estimators are compared with traditional Monte Carlo and polynomial chaos expansion estimates. Specifically, we show consistent sensitivity ranks for both bi‐ (1D/0D) and tri‐fidelity (3D/1D/0D) estimators, and superior variance reduction compared to traditional single‐fidelity Monte Carlo estimators for the same computational budget. As the computational burden of Monte Carlo estimators for Sobol' indices is significantly affected by the problem dimensionality, polynomial chaos expansion is found to have lower computational cost for idealized models with smooth stochastic response. [ABSTRACT FROM AUTHOR]

Sentiment analysis is a fine‐grained analysis task that aims to identify the sentiment polarity of a specified sentence. Existing methods in Chinese sentiment analysis tasks only consider sentiment features from a single pole and scale and thus cannot fully exploit and utilise sentiment feature information, making their performance less than ideal. To resolve the problem, the authors propose a new method, GP‐FMLNet, that integrates both glyph and phonetic information and design a novel feature matrix learning process for phonetic features with which to model words that have the same pinyin information but different glyph information. Our method solves the problem of misspelling words influencing sentiment polarity prediction results. Specifically, the authors iteratively mine character, glyph, and pinyin features from the input comments sentences. Then, the authors use soft attention and matrix compound modules to model the phonetic features, which empowers their model to keep on zeroing in on the dynamic‐setting words in various positions and to dispense with the impacts of the deceptive‐setting ones. Experiments on six public datasets prove that the proposed model fully utilises the glyph and phonetic information and improves on the performance of existing Chinese sentiment analysis algorithms. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Exercise Therapy & Rehabilitation is the property of Yavuz Yakut 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.)

Assessment measures that quantify decision-making abilities in children and adolescents are limited. In the current study, a novel computerized Decision-Making Task (DMT), which identifies the process that is involved in decision-making, was developed based on an existing information-boards paradigm. The overall aim was to validate the DMT in a paediatric TBI population. This prospective study investigated the performance on the DMT for children post-TBI (n = 49; 7–15 years) compared to typically developing controls (n = 22; 7–15 years), and investigated the psychometric properties of the DMT by examining internal consistency-related reliability, convergent validity (measures of decision-making, working memory, functional outcomes, and behaviour), and divergent validity (vocabulary). Significant differences were detected for performance on the DMT between children post-TBI and the control group. Psychometric properties of the DMT were acceptable, with variable findings for convergent validity (working memory, functional outcomes, and behaviour). This is the first study to develop and investigate a novel computerised task to assess decision-making skills in a paediatric TBI population. Results cautiously suggest that the DMT is a valid and a reliable measure of decision-making in our clinical sample. [ABSTRACT FROM AUTHOR]

Detecting and analyzing various defect types in semiconductor materials is an important prerequisite for understanding the underlying mechanisms and tailoring the production processes. Analysis of microscopy images that reveal defects typically requires image analysis tasks such as segmentation and object detection. With the permanently increasing amount of data from experiments, handling these tasks manually becomes more and more impossible. In this work, we combine various image analysis and data mining techniques to create a robust and accurate, automated image analysis pipeline for extracting the type and position of all defects in a microscopy image of a KOH-etched 4H-SiC wafer. [ABSTRACT FROM AUTHOR]

Explainable Artificial Intelligence (XAI) attempts to help humans understand machine learning decisions better and has been identified as a critical component toward increasing the trustworthiness of complex black-box systems, such as deep neural networks. In this article, we propose a generic and comprehensive framework named SNIPPET and create a user interface for the subjective evaluation of visual explanations, focusing on finding human-friendly explanations. SNIPPET considers human-centered evaluation tasks and incorporates the collection of human annotations. These annotations can serve as valuable feedback to validate the qualitative results obtained from the subjective assessment tasks. Moreover, we consider different user background categories during the evaluation process to ensure diverse perspectives and comprehensive evaluation. We demonstrate SNIPPET on a DeepFake face dataset. Distinguishing real from fake faces is a non-trivial task even for humans that depends on rather subtle features, making it a challenging use case. Using SNIPPET, we evaluate four popular XAI methods which provide visual explanations: Gradient-weighted Class Activation Mapping, Layer-wise Relevance Propagation, attention rollout, and Transformer Attribution. Based on our experimental results, we observe preference variations among different user categories. We find that most people are more favorable to the explanations of rollout. Moreover, when it comes to XAI-assisted understanding, those who have no or lack relevant background knowledge often consider that visual explanations are insufficient to help them understand. We open-source our framework for continued data collection and annotation at https://github.com/XAI-SubjEvaluation/SNIPPET. [ABSTRACT FROM AUTHOR]

The development of Web 3.0 is aimed at utilizing available information to enhance users' experiences of searching, allowing this process to be catered by different users to their own needs. However, websites do not make life easier for blind users. Unconditional situations lead to the frustration and total exclusion of blind users. To overcome the accessibility problems, this study aims to develop a blind user mental model (BlUMM) for browsing websites. We conducted a qualitative empirical study via usability test with structured interviews and task analysis in five different scenarios on a website using JAWS screen reader. Six blind users participated in this study. Their feedback was recorded on a video and transcribed into a GOMS model. From the activities, the participants indicated the challenges they faced to accomplish the tasks. The information obtained was analyzed, and BlUMM was developed based on Nielsen's usability principles, GOMS model, and Norman's action cycle model. The methodology was divided into three phases. Phase 1 was the identification of the challenges encountered by blind users while browsing the website. Phase 2 was the analysis of the ability of the blind users to accomplish tasks. Phase 3 involved mapping the blind users' actions. As a result, the BlUMM was developed with 14 elements, which were set/plan the goal, explore the browser/website, apply the existing mental model, discover new features, select features, predict the feedback, execute the plan, execute the task, perceive the feedback, interpret the feedback, change strategy, accomplish the goal, update the strategy, and implement the new mental model with four stages of user action: planning, selection, execution, and evaluation. [ABSTRACT FROM AUTHOR]

Early skin cancer diagnosis saves lives as the disease can be successfully treated through complete excision. Computer-aided diagnosis methods are developed using artificial intelligence techniques to help earlier detection and identify hidden causes leading to cancers in skin lesion images. In skin cancer image classification problems, an ensemble of classifiers has demonstrated better classification ability than a single classification algorithm. Traditionally, training an ensemble uses the complete set of original features, where some of these features can be redundant or irrelevant and hence, may not provide useful information in generating good models for ensemble classification. Moreover, newly created features may help improve classification performance. To address this issue, the existing methods have used feature construction for building an ensemble classifier, which usually creates a fixed number of features that may fit the training data too well, resulting in poor test performance. This study develops a novel classification approach that combines ensemble learning, feature selection, and feature construction utilizing genetic programming (GP) to handle the above limitations. The proposed method automatically evolves variable-length feature vectors consisting of GP-selected and GP-constructed features suitable for training an ensemble classifier. This study evaluates the effectiveness of the proposed method on two benchmark real-world skin image datasets that include dermoscopy and standard camera images. The experimental results reveal that the proposed algorithm significantly outperforms four state-of-the-art convolutional neural network methods, the existing GP approaches, and 11 commonly used machine learning methods. Furthermore, this study also includes interpreting evolved individuals that highlight important skin cancer characteristics playing a vital role in discriminating images of different cancer classes. This study shows that high classification performance can be achieved at a low cost of computational resources and inference time, and accordingly, this method is potentially suitable to be implemented in mobile devices for the automated screening of skin lesions and many other malignancies in low-resource settings. [ABSTRACT FROM AUTHOR]

To date, psychological theories of aggression have shown limited applicability to the first-hand experience of real-world violence. Examining the experiences of those directly involved in violent offending, may enable professionals to develop a strengthened and more refined understanding. The current study aimed to develop a theoretical model of offender experiences of decision-making, during the commission of a violent offence. The Critical Decision Method (CDM) for cognitive task analysis was conducted within a British prison, with nine participants who had been convicted of a serious violent offence. Interview transcripts were analysed using constructivist Grounded Theory (CGT). Seven themes were interpreted from participant narratives: (i) predisposing influences, (ii) environment, (iii) emotion, (iv) triggering events, (v) decision to act, (vi) identity and (vii) severity of harm caused to the victim. Participants' narratives were found to focus upon a process of decision-making which followed a sequence of events, from the impact of early experiences, through to the in-the-moment decisions, that led to acting in a violent manner. This study provides novel insights into the experience of violent decision-making. Participants emphasised the role of emotion and identity. A range of clinical implications are discussed which highlight the important role of emotion regulation and identity transformation in the rehabilitation of violent offenders. [ABSTRACT FROM AUTHOR]

Introduction: This study aimed to describe the fixation location and the time of the longer fixation of expert and novice futsal coaches before the ball was in play in futsal set pieces. Methods: A total of 10 experts (ages 48 ± 5) and 10 novice coaches (ages 40 ± 7) participated in the study. They observed that 38 video clips were created to mimic the attack and defensive set-piece moments of the game. Data were collected in a standardized video analysis task using the pupil invisible eye tracker and processed through the pupil cloud platform. The Mann-Whitney test was conducted to evaluate differences in gaze duration between game moments (attack and defense set pieces) and groups (expert vs. novice). Gaze duration was also compared for gaze location between groups. For further comparisons, the game moments (attack and defense set pieces) and the gaze location were summarized in two-dimensional graphics using correspondence analysis. Results and discussion: The results revealed higher values of gaze duration for attack and defense set pieces for the group of experts than for novices. When considering gaze duration, expert coaches had higher values than novices for the attacker 3, defender 3, barrier 1st, and barrier 2nd gaze locations. The correspondence analysis showed different strategies of visual search and, consequently, gaze locations for attack and defense set pieces. In particular, there was different correspondence for free kicks between the level of expertise and gaze location, while corner and sideline kicks revealed some correspondence between the groups and the gaze location. In free kicks, coaches should be particularly concerned about the relationship between attacker and defender three and the barrier 1st and 2nd line positions. In corner and sideline kicks, coaches should be particularly aware of the relationship between attackers' and defenders' positions. [ABSTRACT FROM AUTHOR]

Aspect Sentiment Quad Prediction is a research topic of paramount significance and complexity within the Aspect-Based Sentiment Analysis task. Leveraging the generative paradigm of the T5 model, we achieve end-to-end extraction of aspect sentiment elements by paraphrasing the original text into sentences predefined by templates. Current research predominantly confines templates to single sentences or directly concatenates sentiment elements using a few symbols, limiting the model's reasoning opportunities. In this work, we introduce a Self-Inference Template (SIT) to guide the model in thoughtful reasoning, facilitating a step-by-step inference generation process. This approach enables the model to more accurately identify aspect sentiment elements and their interdependencies. Experimental results demonstrate a significant improvement in quadruplet prediction performance under constant time costs, effectively mitigating overfitting issues caused by limited data volume to some extent. [ABSTRACT FROM AUTHOR]

Document Layout Analysis, which is the task of identifying different semantic regions inside of a document page, is a subject of great interest for both computer scientists and humanities scholars as it represents a fundamental step towards further analysis tasks for the former and a powerful tool to improve and facilitate the study of the documents for the latter. However, many of the works currently present in the literature, especially when it comes to the available datasets, fail to meet the needs of both worlds and, in particular, tend to lean towards the needs and common practices of the computer science side, leading to resources that are not representative of the humanities real needs. For this reason, the present paper introduces U-DIADS-Bib, a novel, pixel-precise, non-overlapping and noiseless document layout analysis dataset developed in close collaboration between specialists in the fields of computer vision and humanities. Furthermore, we propose a novel, computer-aided, segmentation pipeline in order to alleviate the burden represented by the time-consuming process of manual annotation, necessary for the generation of the ground truth segmentation maps. Finally, we present a standardized few-shot version of the dataset (U-DIADS-BibFS), with the aim of encouraging the development of models and solutions able to address this task with as few samples as possible, which would allow for more effective use in a real-world scenario, where collecting a large number of segmentations is not always feasible. [ABSTRACT FROM AUTHOR]

One potential way to enhance the performance of facial expression recognition (FER) is to augment the training set by increasing the number of samples. By incorporating multiple FER datasets, deep learning models can extract more discriminative features. However, the inconsistent labeling criteria and subjective biases found in annotated FER datasets can significantly hinder the recognition accuracy of deep learning models when handling mixed datasets. Effectively perform joint training on multiple datasets remains a challenging task. In this study, we propose a joint training method for training an FER model using multiple FER datasets. Our method consists of four steps: (1) selecting a subset from the additional dataset, (2) generating pseudo-continuous labels for the target dataset, (3) refining the labels of different datasets using continuous label mapping and discrete label relabeling according to the labeling criteria of the target dataset, and (4) jointly training the model using multi-task learning. We conduct joint training experiments on two popular in-the-wild FER benchmark databases, RAF-DB and CAER-S, while utilizing the AffectNet dataset as an additional dataset. The experimental results demonstrate that our proposed method outperforms the direct merging of different FER datasets into a single training set and achieves state-of-the-art performance on RAF-DB and CAER-S with accuracies of 92.24% and 94.57%, respectively. [ABSTRACT FROM AUTHOR]

The ever-evolving social media discourse has witnessed an overwhelming use of memes to express opinions or dissent. Besides being misused for spreading malcontent, they are mined by corporations and political parties to glean the public's opinion. Therefore, memes predominantly offer affect-enriched insights towards ascertaining the societal psyche. However, the current approaches are yet to model the affective dimensions expressed in memes effectively. They rely extensively on large multimodal datasets for pre-training and do not generalize well due to constrained visual-linguistic grounding. In this paper, we introduce MOOD (Meme emOtiOns Dataset), which embodies six basic emotions. We then present ALFRED (emotion-Aware muLtimodal Fusion foR Emotion Detection), a novel multimodal neural framework that (i) explicitly models emotion-enriched visual cues, and (ii) employs an efficient cross-modal fusion via a gating mechanism. Our investigation establishes ALFRED's superiority over existing baselines by 4.94% F1. Additionally, ALFRED competes strongly with previous best approaches on the challenging Memotion task. We then discuss ALFRED's domain-agnostic generalizability by demonstrating its dominance on two recently-released datasets – HarMeme and Dank Memes, over other baselines. Further, we analyze ALFRED's interpretability using attention maps. Finally, we highlight the inherent challenges posed by the complex interplay of disparate modality-specific cues toward meme analysis. [ABSTRACT FROM AUTHOR]

Most current methods for image emotion analysis suffer from the affective gap, in which features directly extracted from images are supervised by a single emotional label, which may not align with users’ perceived emotions. To effectively address this limitation, this article introduces a novel multi-stage perception approach inspired by the human staged emotion perception process. The proposed approach comprises three perception modules: entity perception, attribute perception, and emotion perception. The entity perception module identifies entities in images, while the attribute perception module captures the attribute content associated with each entity. Finally, the emotion perception module combines entity and attribute information to extract emotion features. Pseudo-labels of entities and attributes are generated through image segmentation and vision-language models to provide auxiliary guidance for network learning. A progressive understanding of entities and attributes allows the network to hierarchically extract semantic-level features for emotion analysis. Comprehensive experiments on image emotion classification, regression, and distribution learning demonstrate the superior performance of our multi-stage perception network. [ABSTRACT FROM AUTHOR]

The advent of Artificial Intelligence (AI) technologies has precipitated the rise of asynchronous video interviews (AVIs) as an alternative to conventional job interviews. These one-way video interviews are conducted online and can be analyzed using AI algorithms to automate and speed up the selection procedure. In particular, the swift advancement of Large Language Models (LLMs) has significantly decreased the cost and technical barrier to developing AI systems for automatic personality and interview performance evaluation. However, the generative and task-unspecific nature of LLMs might pose potential risks and biases when evaluating humans based on their AVI responses. In this study, we conducted a comprehensive evaluation of the validity, reliability, fairness, and rating patterns of two widely-used LLMs, GPT-3.5 and GPT-4, in assessing personality and interview performance from an AVI. We compared the personality and interview performance ratings of the LLMs with the ratings from a task-specific AI model and human annotators using simulated AVI responses of 685 participants. The results show that LLMs can achieve similar or even better zero-shot validity compared with the task-specific AI model when predicting personality traits. The verbal explanations for predicting personality traits generated by LLMs are interpretable by the personality items that are designed according to psychological theories. However, LLMs also suffered from uneven performance across different traits, insufficient test-retest reliability, and the emergence of certain biases. Thus, it is necessary to exercise caution when applying LLMs for human-related application scenarios, especially for significant decisions such as employment. [ABSTRACT FROM AUTHOR]

The emotional content of several databases are annotated with continuous-time (CT) annotations, providing traces with frame-by-frame scores describing the instantaneous value of an emotional attribute. However, having a single score describing the global emotion of a short segment is more convenient for several emotion recognition formulations. A common approach is to derive sentence-level (SL) labels from CT annotations by aggregating the values of the emotional traces across time and annotators. How similar are these aggregated SL labels from labels originally collected at the sentence level? The release of the MSP-Podcast (SL annotations) and MSP-Conversation (CT annotations) corpora provides the resources to explore the validity of aggregating SL labels from CT annotations. There are 2,884 speech segments that belong to both corpora. Using this set, this study (1) compares both types of annotations using statistical metrics, (2) evaluates their inter-evaluator agreements, and (3) explores the effect of these SL labels on speech emotion recognition (SER) tasks. The analysis reveals benefits of using SL labels derived from CT annotations in the estimation of valence. This analysis also provides insights on how the two types of labels differ and how that could affect a model. [ABSTRACT FROM AUTHOR]

Facial expressions are important nonverbal behaviors that humans use to express their feelings. Clinical research have shown that depressed patients have poor facial expressiveness and mimicry. As a result, we propose a VFEM experiment with seven expressions to explore variations in facial expression features between depressed patients and normal people, including anger, disgust, fear, happiness, neutrality, sadness, and surprise. It has been discovered through VFEM experiments that depressed patients frequently exhibit negative facial expressions. Meanwhile, we propose a depression facial expression recognition (Dep-FER) model in this research. Dep-FER involves three innovative and crucial components: Mask Multi-head Self-Attention (MMSA), facial action unit similarity loss function (AUs Loss), and case-control loss function (CC Loss). MMSA can filter out disturbing samples and force to learn the relationship between different samples. AUs Loss utilizes the similarity between each expression AU and the model output to improve the generalization ability of the model. CC Loss addresses the intrinsic link between the depressed and normal patient categories. Dep-FER achieves excellent performance in VFEM and outperforms existing comparative models. [ABSTRACT FROM AUTHOR]

As a fine-grained and local expression behavior measurement, facial action unit (FAU) analysis (e.g., detection and intensity estimation) has been documented for its time-consuming, labor-intensive, and error-prone annotation. Thus a long-standing challenge of FAU analysis arises from the data scarcity of manual annotations, limiting the generalization ability of trained models to a large extent. Amounts of previous works have made efforts to alleviate this issue via semi/weakly supervised methods and extra auxiliary information. However, these methods still require domain knowledge and have not yet avoided the high dependency on data annotation. This article introduces a robust facial representation model MAE-Face for AU analysis. Using masked autoencoding as the self-supervised pre-training approach, MAE-Face first learns a high-capacity model from a feasible collection of face images without additional data annotations. Then after being fine-tuned on AU datasets, MAE-Face exhibits convincing performance for both AU detection and AU intensity estimation, achieving a new state-of-the-art on nearly all the evaluation results. Further investigation shows that MAE-Face achieves decent performance even when fine-tuned on only 1% of the AU training set, strongly proving its robustness and generalization performance. The pre-trained model is available at our GitHub repository. [ABSTRACT FROM AUTHOR]

Emotion recognition is a critical component of affective computing. Training accurate machine learning models for emotion recognition typically requires a large amount of labeled data. Due to the subtleness and complexity of emotions, multiple evaluators are usually needed for each affective sample to obtain its ground-truth label, which is expensive. To save the labeling cost, this paper proposes an inconsistency-based active learning approach for cross-task transfer between emotion classification and estimation. Affective norms are utilized as prior knowledge to connect the label spaces of categorical and dimensional emotions. Then, the prediction inconsistency on the two tasks for the unlabeled samples is used to guide sample selection in active learning for the target task. Experiments on within-corpus and cross-corpus transfers demonstrated that cross-task inconsistency could be a very valuable metric in active learning. To our knowledge, this is the first work that utilizes prior knowledge on affective norms and data in a different task to facilitate active learning for a new task, even the two tasks are from different datasets. [ABSTRACT FROM AUTHOR]

Emotion recognition in conversation (ERC) based on multiple modalities has attracted enormous attention. However, most research simply concatenated multimodal representations, generally neglecting the impact of cross-modal correspondences and uncertain factors, and leading to the cross-modal misalignment problems. Furthermore, recent methods only considered simple contextual features, commonly ignoring semantic clues and resulting in an insufficient capture of the semantic consistency. To address these limitations, we propose a novel multi-level alignment and cross-modal unified semantic graph refinement network (MA-CMU-SGRNet) for ERC task. Specifically, a multi-level alignment (MA) is first designed to bridge the gap between acoustic and lexical modalities, which can effectively contrast both the instance-level and prototype-level relationships, separating the multimodal features in the latent space. Second, a cross-modal uncertainty-aware unification (CMU) is adopted to generate a unified representation in joint space considering the ambiguity of emotion. Finally, a dual-encoding semantic graph refinement network (SGRNet) is investigated, which includes a syntactic encoder to aggregate information from near neighbors and a semantic encoder to focus on useful semantically close neighbors. Extensive experiments on three multimodal public datasets show the effectiveness of our proposed method compared with the state-of-the-art methods, indicating its potential application in conversational emotion recognition. [ABSTRACT FROM AUTHOR]

Automatic sensing of emotional information in speech is important for numerous everyday applications. Conventional Speech Emotion Recognition (SER) models rely on averaging or consensus of human annotations for training, but emotions and raters’ interpretations are subjective in nature, leading to diverse variations in perceptions. To address this, our proposed approach integrates the rater's subjectivity by forming the Perception-Coherent Clusters (PCC) of raters to be used to derive expanded label space for learning to improve SER. We evaluate our method on the IEMOCAP and the MSP-Podcast corpora, considering scenarios of fixed and variable raters, respectively. The proposed architecture, Rater Perception Coherency (RPC)-based SER surpasses single-task models with consensus labels by achieving UAR improvements of 3.39% for the IEMOCAP and 2.03% for the MSP-Podcast. Further analysis provides comprehensive insights into the contributions of these perception consistency clusters in SER learning. [ABSTRACT FROM AUTHOR]

Emergent states are temporal group phenomena that arise from collective affective, behavioral, and cognitive processes shared among the group's members during their interactions. Cohesion is one such state, mainly conceptualized by scholars as affective in nature, and frequently distinguished into the two dimensions social and task cohesion. Whereas social cohesion is related to the need of belonging to a group, task cohesion is related to the group's goals and tasks. In this article, we emphasize the importance of behavioral interaction dynamics to predict cohesion's dynamics. Drawing from Social Science insights, we investigate the interplay between social and task cohesion to predict their dynamics across group tasks from nonverbal behavioral features. Three computational architectures exploiting transfer learning are presented. Transfer learning capitalizes on information learnt by a model for a specific dimension to predict the dynamics of the other dimension. Results show that integrating the influence of social cohesion to predict dynamics of task cohesion outperforms state-of-the-art. To predict dynamics of social cohesion, a model integrating the reciprocal impact of social and task cohesion significantly improves performance with respect to the state-of-the-art and a model only integrating the impact of task cohesion on dynamics of social cohesion. [ABSTRACT FROM AUTHOR]

Computer mouse tracking is a simple and cost-efficient way to gather continuous behavioral data. As theory suggests a relationship between affect and sensorimotor processes, the computer mouse might be usable for affect sensing. However, the processes underlying a connection between mouse usage and affect are complex, hitherto empirical evidence is ambiguous, and the research area lacks longitudinal studies. The present work brings forward a longitudinal field study in which 179 participants hourly self-reported their affect while their mouse usage was tracked both during their self-directed, contextless as well as task-bound computer use over the course of 14 days, resulting in a dataset comprising 10,760 instances of data collection. Extensive statistical analysis using null hypothesis significance testing and machine learning reveal weak and sporadic relationships between mouse usage and longitudinal self-reported affect at best. The results of this study challenge the use of computer mouse tracking for longitudinal affect sensing and point to a necessity for more research. [ABSTRACT FROM AUTHOR]

Multi-modal classifiers for emotion recognition have become prominent, as the emotional states of subjects can be more comprehensively inferred from Electroencephalogram (EEG) signals and eye movements. However, existing classifiers experience a decrease in performance due to the distribution shift when applied to new users. Unsupervised domain adaptation (UDA) emerges as a solution to address the distribution shift between subjects by learning a shared latent feature space. Nevertheless, most UDA approaches focus on a single modality, while existing multi-modal approaches do not consider that fine-grained structures should also be explicitly aligned and the learned feature space must be discriminative. In this paper, we propose Coarse and Fine-grained Distribution Alignment with Correlated and Separable Features (CFDA-CSF), which performs a coarse alignment over the global feature space, and a fine-grained alignment between modalities from each domain distribution. At the same time, the model learns intra-domain correlated features, while a separable feature space is encouraged on new subjects. We conduct an extensive experimental study across the available sessions on three public datasets for multi-modal emotion recognition: SEED, SEED-IV, and SEED-V. Our proposal effectively improves the recognition performance in every session, achieving an average accuracy of 93.05%, 85.87% and 91.20% for SEED; 85.72%, 89.60%, and 86.88% for SEED-IV; and 88.49%, 91.37% and 91.57% for SEED-V. [ABSTRACT FROM AUTHOR]

EEG emotion recognition has been hampered by the clear individual differences in the electroencephalogram (EEG). Nowadays, domain adaptation is a good way to deal with this issue because it aligns the distribution of data across subjects. However, the performance for EEG emotion recognition is limited by the existing research, which mainly focuses on the global alignment between the source domain and the target domain and ignores much fine-grained information. In this study, we propose a method called Graph-based Unsupervised Subdomain Adaptation (Gusa), which simultaneously aligns the distribution between the source and target domains in a fine-grained way from both the channel and emotion subdomains. Gusa employs three modules, such as the Node-wise Domain Constraints Module to align each EEG channel and obtain a domain-variant representation, the Class-level Distribution Constraints Module, and the Emotion-wise Domain Constraints Module, to collect more fine-grained information, create more discriminative representations for each emotion, and lessen the impact of noisy emotion labels. The studies on the SEED, SEED-IV, and MPED datasets demonstrate that Gusa significantly improves the ability of EEG to recognize emotions and can extract more granular and discriminative representations for EEG. [ABSTRACT FROM AUTHOR]

Recently, multi-modal affective computing has demonstrated that introducing multi-modal information can enhance performance. However, multi-modal research faces significant challenges due to its high requirements regarding data acquisition, modal integrity, and feature alignment. The widespread use of multi-modal pre-training methods offers the possibility of aiding visual sentiment analysis by introducing cross-domain knowledge. This paper proposes a Multi-Aspect Semantic Auxiliary Network (MASANet) for visual sentiment analysis. Specifically, MASANet achieves modality expansion through cross-modal generation, making it possible to introduce cross-domain semantic assistance. Then, a cross-modal gating module and an adaptive modal fusion module are proposed for aspect-level and cross-modal interaction, respectively. In addition, a designed semantic polarity constraint loss is presented to improve sentiment multi-classification performance. Evaluations of eight widely-used affective image datasets demonstrate that our proposed method outperforms the state-of-the-art methods. Further ablation experiments and visualization results also confirm the effectiveness of the proposed method and its modules. [ABSTRACT FROM AUTHOR]