Abstract Finding students at high risk of poor academic performance as early possible plays an important role in improving education quality. To do so, most existing studies have used the traditional machine learning algorithms to predict students’ achievement based on their behavior data, from which features are extracted manually thanks expert experience and knowledge. However, owing increase varieties overall volume behavioral it has become more challenging identify high-quality...

With the advancement of information technology, Social Internet Things (SIoT) has fostered integration physical devices and social networks, deepening study complex interaction patterns. Text Attribute Graphs (TAGs) capture both topological structures semantic attributes, enhancing analysis interactions within SIoT. However, existing graph learning methods are typically designed for complete attributed graphs, common issue missing attributes in Missing (AMGs) increases difficulty tasks. To...

Graph neural networks (GNNs) have been widely used for graph structure learning and achieved excellent performance in tasks such as node classification link prediction. Real-world imply complex various semantic information are often referred to heterogeneous (HINs). Previous GNNs laboriously modeled with pairwise relations, which the representation is incomplete severely hinders embedded learning. Therefore, conventional cannot satisfy demand discovery HINs. In this article, we propose an...

The exploration of self-supervised information mining heterogeneous datasets has gained significant traction in recent years. Heterogeneous graph neural networks (HGNNs) have emerged as a highly promising method for handling (HINs) due to their superior performance. These leverage aggregation functions convert pairwise relations-based features from raw graphs into embedding vectors. However, real-world HINs contain valuable higher-order relations that are often overlooked but can provide...

Abstract Spatial transcriptomics reveals the spatial distribution of genes in complex tissues, providing crucial insights into biological processes, disease mechanisms, and drug development. The prediction gene expression based on cost-effective histology images is a promising yet challenging field research. Existing methods for from exhibit two major limitations. First, they ignore intricate relationship between cell morphological information expression. Second, these do not fully utilize...

Attribute graphs are a crucial data structure for graph communities. However, the presence of redundancy and noise in attribute can impair aggregation effect integrating two different heterogeneous distributions structural features, resulting inconsistent distorted that ultimately compromises accuracy reliability learning. For instance, redundant or irrelevant attributes result overfitting, while noisy lead to underfitting. Similarly, features affect representations, making it challenging...

Describing an object from multiple perspectives often leads to incomplete data representation. Consequently, learning consistent representations for missing views has emerged as a key focus in the realm of Incomplete Multi-view Representation Learning (IMRL). In recent years, various strategies such subspace learning, matrix decomposition, and deep have been harnessed develop numerous IMRL methods. this paper, our primary research revolves around IMRL, with particular emphasis on addressing...

Community detection aims to identify dense subgroups of nodes within a network. However, in real-world networks, node attributes are often missing, making traditional methods less effective. In networks with missing attributes, the main challenge community is deal attribute information efficiently and use network structure make accurate predictions. This article proposes an innovative method called contrastive sampling-aggregating transformer (CSAT) for attribute-missing networks. CSAT...

Phragmites australis straw (PAS) is an abundant and renewable wetland lignocellulose. Bacillus coagulans IPE22 a robust thermophilic strain with pentose-utilizing capability excellent resistance to growth inhibitors. This work focused on the process study of lactic acid (LA) production from P. lignocellulose which has not been attempted previously. By virtue feature IPE22, two fermentation processes (i.e., separated integrated process), were developed compared under non-sterilized...

Social link is an important index to understand master students’ mental health and social ability in educational management. Extracting hidden strength from rich daily life behaviors has also become attractive research hotspot. Devices with positioning functions record many spatiotemporal behavior data, which can infer links. However, under the guidance of school regulations, activities have a certain regularity periodicity. Traditional methods usually compare co-occurrence frequency two...

Abstract The biological properties and synthesis of ferruginol as a classical abietane-type diterpenoid with an aromatic C ring are reviewed. A strategy overview from 1954 to 2023 toward the total may provide some references for future design new diterpenoids natural products. 1 Introduction 2 Biological Activity Ferruginol 3 Strategies Total Synthesis 3.1 Bogert–Cook 3.2 Robinson Annulation 3.3 Domino 3.4 Intramolecular Friedel–Crafts Alkylation 3.5 Oxidative Free-Radical Cyclization 3.6...

For autonomous driving, accurate trajectory prediction is paramount, necessitating effective harnessing of spatiotemporal data. This study proposes an innovative Spatiotemporal Transformer-based model, enhancing precision by leveraging a multi-head self-attention mechanism. mechanism intricately captures both inter-vehicular interactions and temporal dependencies. The structured around LSTM-based encoder-decoder framework, innovatively considers spatial among observed future trajectories...

ABSTRACT Spatial transcriptomics reveals the spatial distribution of genes in complex tissues, providing crucial insights into biological processes, disease mechanisms, and drug development. The prediction gene expression based on cost-effective histology images is a promising yet challenging field research. Existing methods for from exhibit two major limitations. First, they ignore intricate relationship between cell morphological information expression. Second, these do not fully utilize...

With the development of intelligent connected vehicle technology, human-machine shared control has gained popularity in following due to its effectiveness driver assistance. However, traditional systems struggle maintain stability when reaction time fluctuates, as these variations require different levels system intervention. To address this issue, proposed assistance (HM-VFAS) integrates outputs under various states with system. The employs an model that accounts for delays, simulating...

Accurately estimating time of arrival (ETA) for trucks is crucial optimizing transportation efficiency in logistics. GPS trajectory data offers valuable information ETA, but challenges arise due to temporal sparsity, variable sequence lengths, and the interdependencies among multiple trucks. To address these issues, we propose Temporal-Attribute-Spatial Tri-space Coordination (TAS-TsC) framework, which leverages three feature spaces-temporal, attribute, spatial-to enhance ETA. Our framework...

Attribute graphs are ubiquitous in multimedia applications, and graph representation learning (GRL) has been successful analyzing attribute data. However, incomplete data missing node attributes can have a negative impact on media knowledge discovery. Existing methods for handling limited assumptions or fail to capture complex attribute-graph dependencies. To address these challenges, we propose Graph Contrastive Learning (AmGCL), framework AmGCL leverages Dirichlet energy minimization-based...

Due to the poor quality of social responsibility information disclosure Chinese enterprises, coupled with impact spread COVID-19, uncertainty global economic environment and policies has increased significantly, development enterprises is facing unprecedented challenges.Based on sample A-share listed companies in Shanghai Shenzhen from 2010 2019, this study examined corporate performance moderating role policy uncertainty. Research shows that can improve by disclosing high-quality...

Despite the widespread use of graph structures in real world, problem missing attributes on graphs still exists. Recently, researchers have proposed a series inductive representation learning methods that can be used conjunction with deep techniques. Yet, attribute-missing currently lacks an effective paradigm. In this study, Contrastive Sampling-Aggregation Transformer (CSAT) is introduced for representations. CSAT utilizes Graph Sampling Auto-Encoder (CGS-AE) to aggregate structural and...

Clustering is a fundamental and hot issue in the unsupervised learning area. With rapid development of deep graph neural networks (GNNs) techniques, researchers have proposed series effective clustering methods. However, most existing approaches adopt conventional to aggregate neighborhood information, where only pairwise relations are considered. Moreover, redundancy/noise raw data samples may result less accurate sample inferior results. In this paper, we new GNNs based method, which...