Cross-subject motor imagery classification remains challenging due to EEG data scarcity and inter-subject variability. This study proposes a novel framework integrating generative data augmentation with domain adaptation. First, we employ a diffusion probabilistic model to generate high-fidelity synthetic EEG samples, effectively enriching the training data. Subsequently, we propose the AMSC-DANN architecture, which synergizes an Adaptive Multi-Scale Convolution (AMSC) module for extracting multi-granular features with a Domain Adversarial Neural Network (DANN). This combination enables the model to learn discriminative temporal-spectral representations while simultaneously aligning feature distributions across different subjects. Extensive experiments on BCI Competition IV datasets 2a and 2b demonstrate that our proposed framework outperforms state-of-the-art baselines, validating its effectiveness in enhancing cross-subject generalization.

The concept of Negative Hesitation Fuzzy Sets (NHFSs) has been proposed recently. NHFSs show superiority in pattern recognition. In this paper, we propose different distance measures for NHFSs. The distance measures reflect the relationship between the various patterns constructed by NHFSs. The distance measures of the NHFSs can become the basis for creating the similarity measure. Meanwhile, the distance measures can be applied to feature extraction and selection in Machine Learning and Neural Networks when NHFSs represent the data in future work. Illustrative experiments are used to evaluate these distance measures. Failure Examples of different distance measures in pattern recognition problems will be demonstrated and explained. The comparison between different distance measures is also concluded. Distance measures are also applied to the Motor Imaginary Electroencephalogram (EEG) classification task, and the precision has been calculated.

The K-means algorithm utilizes the Euclidean distance metric to quantify the similarity between data points and clusters, with the fundamental objective of assessing the relationship between points. It is important to note that, during the process of clustering, the relationships between the remaining points in the cluster and the points to be measured are ignored. In consideration of the aforementioned issues, this paper proposes the utilization of extension distance for the purpose of evaluating the relationship between the points to be measured and the cluster classes. Furthermore, it introduces a variant of the K-means algorithm based on the separator distance. Through a series of comparative experiments, the effectiveness of the proposed algorithm for clustering fan-shaped datasets is preliminarily verified.

Artificial intelligence has brought tremendous convenience to human life in various aspects. However, during its application, there are still instances where AI fails to comprehend certain problems or cannot achieve flawless execution, necessitating more cautious and thoughtful usage. With the advancements in EEG signal processing technology, its integration with AI has become increasingly close. This idea of interpreting electroencephalogram (EEG) signals illustrates researchers’ desire to explore the deeper relationship between AI and human thought, making human-like thinking a new direction for AI development. Currently, AI faces several core challenges: it struggles to adapt effectively when interacting with an uncertain and unpredictable world. Additionally, the trend of increasing model parameters to enhance accuracy has reached its limits and cannot continue indefinitely. Therefore, this paper proposes revisiting the history of AI development from the perspective of “anthropomorphic computing”, primarily analyzing existing AI technologies that incorporate structures or concepts resembling human brain thinking. Furthermore, regarding the future of AI, we will examine its emerging trends and introduce the concept of “Cyber Brain Intelligence”—a human-like AI system that simulates human thought processes and generates virtual EEG signals.

The Intuitionistic Fuzzy C-means (IFCM) Clustering Algorithm is an extension of the Fuzzy C-means (FCM) Clustering Algorithm. Traditional research concerning the IFCM has predominantly concentrated on the integration of novel distance measures within the objective function. However, the weights in the objective function have consistently adhered to the constraints imposed by the original FCM algorithm. In this study, we propose a Bi-environmental Intuitionistic Fuzzy C-means (Bi-IFCM) Clustering Algorithm, wherein the weights are determined by employing intuitionistic fuzzy logic. The effectiveness of the Bi-IFCM algorithm is validated by its application to four UCI datasets within clustering tasks, demonstrating its superiority relative to previously modified IFCM algorithms. Furthermore, in the image segmentation task using the Berkeley Segmentation Dataset and Benchmark 500 (BSDS500), Bi-IFCM exhibited enhanced performance over preceding IFCM algorithms, as evidenced by improved dice similarity coefficient, accuracy, precision, and recall metrics.

In this article, we reported a novel granulation method composed of complexity information based on permutation entropy (PeEn). This method aims to recognize the electroencephalography (EEG) patterns using this proposed granulation method. First, we define the complexity information for granular computing by a technique with fast calculation, i.e., PeEn. Then, the information granule can be constructed based on the time domain information, which completes complexity information. Together with the support vector machine algorithm, the proposed granulation method outperformed the existing classification methods in accuracy. It is utilized by classifying three motor imaginary EEG signals. Two of them are binary-class datasets, i.e., one dataset includes two-hand actions, and another includes hand and foot actions. The third dataset is multiclass, including two hands and two feet actions. In addition, the proposed granulation method overcomes the difficulties in cross-individual cases when classifying the EEG signals with a higher accuracy than the existing methods. Meanwhile, this classification procedure makes it interpretable and has a high performance.

A novel score function based on the Poincaré metric is proposed and applied to a decision-making problem. Decision-making on Fuzzy Sets (FSs) has been considered due to the flexibility of the data, and it is applied to the decision-making. However, decisions with FSs are sometimes nondecisive even for different membership degrees. Hence, Intuitionistic Fuzzy Sets (IFSs) data is applied to design a score function for the decision-making with the Poincaré metric. This function is supported by the profound information of IFSs; IFSs include hesitation degree together with membership and non-membership degree. Hence, IFS membership and non-membership degree are expressed as two-dimensional vectors satisfying the Poincaré metric for simplification. At the same time, the proposed approach addresses the hesitation information in the IFS data. Next, a score function is proposed, constructed and provided. The proposed score function has a strict monotonic property and addresses the preference without resorting to the accuracy function. The strict monotonic property guarantees the preference of all attributes. Additionally, the existing problem of score function design in IFSs is addressed: they return zero scores even with different meanings for the same membership and non-membership degree. The advantages of the proposed score function over existing ones are demonstrated through illustrative examples. From the calculation results, the proposed decision score function discriminates between all candidates. Hence, the proposed research provides a solid foundation for the hesitation analysis on the decision-making problem.

Virtual Reality (VR), Augmented Reality (AR), Mixed Reality (MR), digital twin, Metaverse and other related digital technologies have attracted much attention in recent years. These new emerging technologies are changing the world significantly. This research introduces a fusion model, i.e. Fusion Universe (FU), where the virtual, physical, and cognitive worlds are merged together. Therefore, it is crucial to establish a set of principles for the fusion model that is compatible with our physical universe laws and principles. This paper investigates several aspects that could affect immersive and interactive experience; and proposes the fundamental principles for Fusion Universe that can integrate physical and virtual world seamlessly.

The initial concept of Negative Hesitation Fuzzy Sets (NHFSs) has been introduced recently. NHFSs are applied to decision-making problems accompanied by soft set theory. In this paper, a detailed clarification of NHFSs is proposed. Meanwhile, we introduced the way to construct membership, non-membership, and negative hesitation degrees by studying the overlap area between the projections of the element and classes in a two-dimensional space. This unified construction has concluded the relationship between NHFSs and Intuitionistic Fuzzy Sets (IFSs). A corollary of cosine similarity satisfying the NHFSs is employed for the pattern recognition problems. Classification of both synthetic numerical examples and the EEG signals are evaluated for the effectiveness of NHFSs in this paper.

The purpose of aspect-category sentiment classification (ACSC) is to determine the sentiment polarity of the predefined aspect category from the texts. Current methods for ACSC have two main limitations. Since the aspect categories are not presented in the given texts, the establishment of relation between the aspect-category and its sentiment opinion expression is challenging using the widely-applied aspect-term sentiment classification approaches. Besides, the aspect-category-related information on document level are ignored during processing. In this work, we focus on dealing with the part-of-speech information based on gated-activation functions. Furthermore, two graph attention networks (GANs) are employed to exploit the document-level sentiment of both the entity and the attribute (intra-entity sentiment tendency and intra-attribute sentiment tendency). The aspect-category detection (ACD) is taken as a auxiliary task to capture the relevant semantic information. Besides, contrastive learning is receiving an increasing amount of interest due to its success in self-supervised representation learning in the field of NLP. By performing contrastive learning, representations of positive examples are drawn closer while those of negative samples are distanced. Comparing with the baseline methods, experimental results reveal that our model achieves the state-of-the-art performance in ACSC tasks.

This paper proposes a novel preference scale function based on the Poincare metric for decision-making with Intuitionistic Fuzzy Sets (IFSs). We first introduce a pair of two-dimensional vectors expressing the IFSs in multi-criteria decision-making problems, which satisfy the Poincare metric, and then construct a preference scale function based on it. The proposed scale function can address, without resorting to the accuracy function, the issue of the existing score function for IFSs returning zero scores when membership and non-membership degrees of an element are the same. The advantages of the proposed scale function over the existing ones are demonstrated through illustrative examples.

3-D human pose estimation or human tracking has always been the focus of research in the human–computer interaction community. As the calibration step of human pose estimation, subject-specific modeling is crucially important to the subsequent pose estimation process. It not only provides a priori knowledge but also clearly defines the tracking target. This article presents a fully automatic subject modeling framework to reconstruct human pose, shape, as well as the body texture in a challenging optimization scenario. By integrating powerful differentiable rendering into the subject-specific modeling pipeline, the proposed method transforms the texture reconstruction problem into analysis by synthesis minimization and solves it efficiently by a gradient-based method. Furthermore, a novel covariance matrix adaptation annealing algorithm is proposed to attack the high-dimensional multimodal optimization problem in an adaptive manner. The domain knowledge of hierarchical human anatomy is seamlessly injected to the annealing optimization process by using a soft covariance matrix mask. All together contributes to the novel algorithm robust to the temptation of local minima. Experiments on the Human3.6 M dataset and the People-Snapshot dataset demonstrate the competitive results to the state of the art both qualitatively and quantitatively.

Multimodality sentiment classification of social media attracts increasing attention, whose main purpose is to predict the sentiment of the target mentioned in the posts. Current research mainly focuses on integrating the multimodal data, but fails to consider the impacts on the target. In this work, we tend to propose a target-oriented multimodal sentiment classification model. Specifically, our model starts with exploiting the target-oriented topic within the text. Then, a multi-head attention network is established to learn the multimodal interaction among textual, visual and topic information, based on which the target-oriented representations of the topic, the text and the image are obtained. Moreover, a gating unit to fuse the multimodal information is also built up. On the task of target-oriented multimodal sentiment classification, experiments on multimodal samples are carried out on manually annotated the dataset. Experimental results reveal that our method significantly reduces the gap over each given target, which sets a foundation to achieve the state-of-arts sentiment classification results.

The development of big data technologies, which have been applied extensively in various areas, has become one of the key factors affecting modern society, especially in the virtual reality environment. This paper provides a comprehensive survey of the recent developments in big data technologies, and their applications to virtual reality worlds, such as the Metaverse, virtual humans, and digital twins. The purpose of this survey was to explore several cutting-edge big data and virtual human modelling technologies, and to raise the issue of future trends in big data technologies and the Metaverse. This survey investigated the applications of big data technologies in several key areas-including e-health, transportation, and business and finance-and the main technologies adopted in the fast-growing virtual world sector, i.e., the Metaverse.

Brainstorming is a widely used problem-solving method that generates a large number of innovative ideas by guiding and stimulating intuitive and divergent thinking. However, in practice, the method is limited by the human brain's capacity or special capabilities, especially by the experience and knowledge they possess. How does our brain create ideas like storming? Based on the new discipline of Extenics, the authors propose a new model that explores the process of how ideas are created in our brain, with the goal of helping people think multi-dimensionally and getting more ideas. With the support of information technology and artificial intelligence, we can systematically collect more information and knowledge than ever before to form a basic-element information base and build human-computer interaction models, to make up for the lack of information and knowledge in the human brain. In addition, the authors provide a methodology to help people think positively in a multidimensional way based on the guidance of Extenics in the brainstorming process.

Oxidative stress has close relationship with the progression of osteoarthritis and thus is an important therapeutic target. The use of antioxidants alone has common disadvantages such as low bioavailabil-ity, poor stability, and rapid joint clearance or toxicity at high concentrations. To achieve both reactive oxygen species (ROS) responsive scavenging ability and drug targeted delivery ability, a novel polymer (PEG-PTK-PEG) was synthesized through a simple and direct reaction between polythioketal (PTK) and m-PEG-acrylate. In addition to excellent mechanical properties and high drug loading capacity, the degra-dation products of PEG-PTK-PEG responding to the ROS accumulating microenvironment also have excel-lent biocompatibility and extremely low toxicity. Further, the PEG-PTK-PEG nanoparticles (NPs) loaded with astaxanthin (ASTA), which is an antioxidant with high biological safety, could effectively reduce ROS expression and release ASTA slowly and accurately in the area of high ROS expression as a result of the polymer degradation. In the OA model, the PEG-PTK-PEG@ASTA could effectively inhibit the expression of ROS, thereby downregulating the MAPK signaling pathway and inducing the conversion of macrophages from the M1 phenotype to the M2 phenotype. The synergistic effect of PEG-PTK-PEG's ROS-responsive drug delivery ability and ROS scavenging ability significantly promoted the therapeutic effect of OA, pro-viding a new and effective strategy for the clinical treatment of OA. (c) 2022 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license ( http://creativecommons.org/licenses/by-nc-nd/4.0/ )

This paper proposes a VR supermarket with an intelligent recommendation. It consists of three parts: VR supermarket, recommendation system, and database. The VR supermarket provides a 360-degree virtual environment for users to move and interact in the virtual environment through VR devices. The recommendation system will make recommendations to the target users based on the data in the database. The intelligent recommendation system is developed based on item similarity (ICF), which solves the cold start problem of ICF. This allows VR supermarkets to present real-time recommendations in any situation. The VR supermarket not only makes up for the lack of user perception of item attributes in traditional online shopping system but also improve user shopping efficiency through an intelligent recommendation system. The application can be extended to enterprise-level systems and can add behavioral and voice interaction with NPC, adding new possibilities for users to do VR shopping at home.

The purpose of this study is to explore the noninvasive human-computer interaction methods that have been widely used in various fields, especially in the field of robot control. To have a deep understanding of the development of the methods, this paper employs “Mapping Knowledge Domains” (MKDs) to find research hotspots in the area to show the future potential development. Through the literature review, this paper found that there was a paradigm shift in the research of noninvasive BCI technologies for robotic control, which has occurred from early 2010 since the rapid development of machine learning, deep learning, and sensory technologies. This study further provides a trend analysis that the combination of data-driven methods with optimized algorithms and human-sensory-driven methods will be the key areas for the future noninvasive method development in robotic control. Based on the above findings, the paper provides a potential developing way of noninvasive HCI methods for related areas including health care, robotic system, and media.

The advances in the electronic age have made the generation of data faster. However, the speed of data analysis has not kept pace with the growth of data. So people try to transform the large data into small-scale data to improve the efficiency of data analysis. Like data dimension reduction, data sampling and data compression are the most commonly used methods at present. The common sampling methods, including random sampling, stratified sampling, have been utilized extensively in various applications. Furthermore, data dimension reduction methods have been deployed in various areas.This paper proposes an optimized high dimension data reduction method based on covariance. Firstly, calculate the covariance matrix of the original dataset and the sub dataset. Then, create a transformation matrix according to the covariance matrix and use this transformation matrix to get a new dataset. Finally, obtain a new sub dataset by adjusting data from sub dataset and new dataset. This method can generate representative datasets and provide effective solutions for big data processing.

Selective exhaustion of over-expressed reactive oxygen species (ROS) is of great significance in the therapy of osteoarthritis (OA) because of the inhibiting effect on oxidative stress and inflammation. Herein, a ROS-scavenging and drug-release platform was prepared via encapsulating dexamethasone acetate (DA)-loaded ROS erasable poly(ethylene glycol)-b-polythioketal-b-poly(ethylene glycol) (PEG-PTK-PEG) micelles (PDM) into an injectable hydrogel. The hydrogel (HDH@PDM) was constructed by Schiff base reaction between hydrazide-grafted hyaluronic acid (HA-ADH) and aldehyde-modified dextran (Dex-ALH), achieving a self-healing property for viscosupplementation. The PDM imparted enhanced antioxidant capability to the hydrogel, which, in turn, endowed the PDM with prolonged retention and sustained DA release. The intraarticularly administered multifunctional injectable hydrogel potently diminished inflammation via depleting ROS and suppressing inflammatory cytokines, as well as downregulating pro-inflammatory M1 macrophages ratio in a rat OA model. The developed therapeutic system significantly alleviated OA symptoms, embodying the excellent capability of preventing cartilage extracellular matrix degeneration with negligible toxicity in vivo. (C) 2021 Elsevier Ltd. All rights reserved.