Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks461 from public data
- A Comprehensive Survey on Graph Anomaly Detection With Deep Learning898
A taxonomy that follows a task-driven strategy and categorizes existing work according to the anomalous graph objects that they can detect is provided, to create a “one-stop-shop” that provides a unified understanding of the problem categories and existing approaches.
- A comprehensive survey on pretrained foundation models: a history from BERT to ChatGPT557
This survey aims to shed light on the research of the PFMs on scalability, security, logical reasoning ability, cross-domain learning ability, cross-domain learning ability, and user-friendly interactive ability for artificial general intelligence.
- 460
- Infrared and visible image fusion via detail preserving adversarial learning456
This paper proposes an end-to-end model for infrared and visible image fusion based on detail preserving adversarial learning that is able to overcome the limitations of the manual and complicated design of activity-level measurement and fusion rules in traditional fusion methods.
- Tri-party deep network representation439
TriDNR is a tri-party deep network representation model, using information from three parties: node structure, node content, and node labels (if available) to jointly learn optimal node representation, which results in up to 79% classification accuracy gain, compared to state-of-the-art methods.
- 416
- A Comprehensive Survey on Automatic Knowledge Graph Construction401
A comprehensively survey of more than 300 methods to summarize the latest developments in knowledge graph construction and presents the paradigms to compare the distinction among these methods along the axis of the data environment, motivation, and architecture.
- A Survey of Community Detection Approaches: From Statistical Modeling to Deep Learning397
A unified architecture of network community-finding methods to characterize the state-of-the-art of the field of community detection is developed and a new taxonomy that divides the existing methods into two categories, namely probabilistic graphical model and deep learning is introduced.
- 268
- 267
- Graph Structure Learning with Variational Information Bottleneck249
VIB-GSL is the first attempt to advance the Information Bottleneck principle for graph structure learning, providing a more elegant and universal framework for mining underlying task-relevant relations.
- A Comprehensive Survey on Deep Clustering: Taxonomy, Challenges, and Future Directions247
A novel taxonomy of deep clustering is introduced based on how two fundamental components: the representation learning module and the clustering module interact, specifically through multistage, generative, iterative, and simultaneous approaches.
- SUGAR: Subgraph Neural Network with Reinforcement Pooling and Self-Supervised Mutual Information Mechanism202
A novel hierarchical subgraph-level selection and embedding-based graph neural network for graph classification, namely SUGAR, to learn more discriminative subgraph representations and respond in an explanatory way and a reinforcement pooling mechanism, which improves the generalization ability of the model.
- A Deep Framework for Cross-Domain and Cross-System Recommendations184
DCDCSR framework outperforms the state-of-the-art CDR and CSR approaches in terms of recommendation accuracy and takes into account the rating sparsity degrees of individual users and items in different domains or systems and uses them to guide the DNN training process for utilizing the rating data more effectively.
- eFraudCom: An E-commerce Fraud Detection System via Competitive Graph Neural Networks182
A competitive graph neural networks (CGNN)-based fraud detection system (eFraudCom) to detect fraud behaviors at one of the largest e-commerce platforms, “Taobao”, and demonstrates that the proposed deep framework CGNN is superior to other baselines in detecting fraud behaviors.
- H2-FDetector: A GNN-based Fraud Detector with Homophilic and Heterophilic Connections179
A Graph Neural Network-based Fraud Detector with Homophilic and Heterophilic Interactions (H2-FDetector for short), which apparently outperforms state-of-the-art baselines on two real public benchmark fraud detection tasks.
- Neighborhood-Aware Attentional Representation for Multilingual Knowledge Graphs172
This paper incorporates neighborhood subgraph-level information of entities, and proposes a neighborhood-aware attentional representation method NAEA for multilingual knowledge graphs that significantly and consistently outperforms state-of-the-art entity alignment models.
- 170
- 152
- 149
- 148
- ComGA139
A novel community-aware attributed graph anomaly detection framework (ComGA), designed to improve the performance of anomaly detection, and a tailored deep graph convolutional network (tGCN) to anomaly detection on attributed graphs.
- Vertically Federated Graph Neural Network for Privacy-Preserving Node Classification132
This paper proposes VFGNN, a federated GNN learning paradigm for privacy-preserving node classification task under data vertically partitioned setting, which can be generalized to existing GNN models.
- An unsupervised parameter learning model for RVFL neural network130
A novel unsupervised network parameter learning method for RVFL, named sparse pre-trained random vector functional link (SP-RVFL for short) network, which uses a sparse autoencoder with ℓ1-norm regularization to adaptively learn superior network parameters for specific learning tasks.
- Time series feature learning with labeled and unlabeled data128
A new Semi-Supervised Shapelets Learning model is presented to efficiently learn shapelets by using both labeled and unlabeled time series data in an integrated model that considers the least squares regression, the power of the pseudo-labels, shapelets regularization, and spectral analysis.
- Task-adaptive Neural Process for User Cold-Start Recommendation120
A novel meta-learning recommender called task-adaptive neural process (TaNP), a new member of the neural process family, where making recommendations for each user is associated with a corresponding stochastic process that enables the model to learn the relevance of different tasks and customize the global knowledge to the task-related decoder parameters for estimating user preferences.
- Knowledge-Preserving Incremental Social Event Detection via Heterogeneous GNNs120
A novel Knowledge-Preserving Incremental Heterogeneous Graph Neural Network (KPGNN) for incremental social event detection that leverages the inductive learning ability of GNNs to efficiently detect events and extends its knowledge from previously unseen data.
- 113
- 111
- 109
- RGDAN: A random graph diffusion attention network for traffic prediction106
A Random Graph Diffusion Attention Network (RGDAN) is proposed for traffic prediction that comprises a graph diffusion attention module and a temporal attention module that can adjust its weights by learning from data like a CNN to capture more realistic spatial dependencies.
- Self-adaptive attribute weighting for Naive Bayes classification103
The proposed method, namely AISWNB, uses immunity theory in Artificial Immune Systems to search optimal attribute weight values, where self-adjusted weight values will alleviate the conditional independence assumption and help calculate the conditional probability in an accurate way.
- A Comprehensive Survey of the Key Technologies and Challenges Surrounding Vehicular Ad Hoc Networks98
The current state of play in VANETs development is surveyed, including the key technologies critical to the field, the resource-management and safety applications needed for smooth operations, the communications and data transmission protocols that support networking, and the theoretical and environmental constructs underpinning research and development.
- Active Discriminative Network Representation Learning96
This paper proposes a novel method, termed as ANRMAB, to learn the active discriminative network representations with a multi-armed bandit mechanism in active learning setting, and designs three active learning query strategies based on the networking data and the learned network representations.
- A Survey on Deep Learning Event Extraction: Approaches and Applications95
A new literature classification of current general domain EE research according to the task definition is introduced and the paradigm and models of EE approaches are summarized, and a comprehensive comparison among different approaches is provided.
- Social Recommendation with an Essential Preference Space95
A novel social recommendation framework, called social recommendation with an essential preferences space (SREPS), which simultaneously models the structural information in the social network, the rating and the consumption Information in the recommender system under the capture of essential preference space is proposed.
- SE-GSL: A General and Effective Graph Structure Learning Framework through Structural Entropy Optimization89
A novel sample-based mechanism for restoring the graph structure via node structural entropy distribution, which increases the connectivity among nodes with larger uncertainty in lower-level communities and enhances the robustness towards noisy and heterophily structures.
- 88
- A Knowledge-Aware Attentional Reasoning Network for Recommendation83
A knowledge-aware attentional reasoning network KARN that incorporates the users' clicked history sequences and path connectivity between users and items for recommendation and can predict the clicking probability of the user with respective to a candidate item is proposed.
- Attribute Weighting via Differential Evolution Algorithm for Attribute Weighted Naive Bayes (WNB)83
The experimental results show that the classification accuracy of the new algorithm DE-WNB is much higher than those of the other algorithms used to compare and the obtained classification accuracy is very good with respect to other common WNB classifiers in literature.
- 82
- Advances in Processing, Mining, and Learning Complex Data: From Foundations to Real‐World Applications82
This special issue contributes to the fundamental research in processing, mining, and learning complex data, focusing on the analysis of complex data sources.
- 82
- 81
- 80
- Heterogeneous Graph Attention Networks for Early Detection of Rumors on Twitter78
A meta-path based heterogeneous graph attention network framework is proposed to capture the global semantic relations of text contents, together with the global structure information of source tweet propagations for rumor detection.
- Link Prediction with Signed Latent Factors in Signed Social Networks77
The proposed SLF model considers four types of relationships: positive, negative, neutral and no relationship at all, and links social relationships of different types to the comprehensive, but opposite, effects of positive and negative SLFs.
- Detecting the evolving community structure in dynamic social networks76
Experimental results on synthetic networks and real-world social networks demonstrate that DECS outperforms in both clustering accuracy and smoothness, contrasted with other state-of-the-art methods.
- 75
- Collaborative Social Group Influence for Event Recommendation75
A new Bayesian latent factor model SogBmf is proposed that combines social group influence and individual preference for event recommendation and demonstrates the effectiveness of the proposed method on real-world data sets.
- 74
- 72
- 71
- 71
- Deep spatial–temporal structure learning for rumor detection on Twitter71
A spatial–temporal structure neural network for rumor detection, termed as STS-NN, which treats the spatial structure and the temporal structure as a whole to model the message propagation, and shows that the approach obtains better performance than baselines in both rumor classification and early detection.
- Multi-View Fusion with Extreme Learning Machine for Clustering71
A novel multi-view fusion clustering framework based on an ELM, called MVEC, which exposes the underlying clustering structures embedded within multi-View data with a high degree of accuracy and a simple yet efficient solution to solve the optimization problem within MVEC.
- 69
- Multi-Aspect enhanced Graph Neural Networks for recommendation68
This work learns the aspect-based sentiments from reviews and uses them to construct multiple aspect-aware user-item graphs, thus giving the edge practical meaning, and designs a routing-based fusion mechanism, which adaptively allocates weights to different aspects to realize the dynamic fusion of aspect preferences.
- 68
- Big social network influence maximization via recursively estimating influence spread67
This paper presents a new algorithm to enable greedy algorithms to perform well in big social network influence maximization that recursively estimates the influence spread using reachable probabilities from node to node.
- Heterogeneous Hypergraph Neural Network for Social Recommendation using Attention Network66
A heterogeneous hypergraph neural network for social recommendation is outlined, called Heterogeneous Hypergraph neural network for Social Recommendation using an Attention Network (HHGSA), that incorporates an attention network to address these issues.
- Time-Variant Graph Classification66
A novel time-variant graph classification task is formulated and a new graph feature is proposed, called a graph-shapelet pattern, for learning and classifying time-Variant graphs, which can be regarded as a graphical extension of a shapelet—a class of discriminative features designed for vector-based temporal data classification.
- Memetic Extreme Learning Machine66
A new Memetic Algorithm (MA)-based Extreme Learning Machine (M-ELM for short) is presented, which embeds the local search strategy into the global optimization framework to obtain optimal network parameters.
- 64
- DAGAD: Data Augmentation for Graph Anomaly Detection64
A novel Data Augmentation-based Graph Anomaly Detection (DAGAD) framework for attributed graphs, equipped with three specially designed modules: an information fusion module employing graph neural network encoders to learn representations, a graph data augmentation module that fertilizes the training set with generated samples, and an imbalance-tailored learning module to discriminate the distributions of the minority (anomalous) and majority (normal) classes.
- 62
- Gapformer: Graph Transformer with Graph Pooling for Node Classification61
Gapformer is presented, a method for node classification that deeply incorporates Graph Transformer with Graph Pooling and coarsens the large-scale nodes of a graph into a smaller number of pooling node via local or global graph pooling methods, and then computes the attention solely with the pooling nodes rather than all other nodes.
- Recommendation with Multi-Source Heterogeneous Information61
Experimental results on two real-world data sets demonstrate that CDNE can use network representation learning to boost the recommendation performance.
- Siamese capsule networks with global and local features for text classification59
Siamese capsule networks with global and local features were proposed, used to glean information about the global semantic differences between categories, which could more accurately represent the semantic distance between different categories.
- 56
- 55
- 55
- Instance cloned extreme learning machine55
Experiments and comparisons on 20 UCI data sets, and validations on image and text classification applications, demonstrate that IC-ELM is able to achieve superior results compared to the original ELM algorithm and its variants, as well as several other classical machine learning algorithms.
- 55
- A Comprehensive Survey on Graph Summarization With Graph Neural Networks54
This article presents a comprehensive survey of progress in deep learning summarization techniques that rely on graph neural networks (GNNs), including recurrent GNNs, convolutional GNNs, graph autoencoders, and graph attention networks.
- CurvDrop: A Ricci Curvature Based Approach to Prevent Graph Neural Networks from Over-Smoothing and Over-Squashing52
A new Curvature-based topology-aware Dropout sampling technique named CurvDrop is proposed, in which the Discrete RicciCurvature is integrated into graph neural networks to enable more expressive graph models.
- SODE: Self-Adaptive One-Dependence Estimators for classification52
This paper proposes a self- Adaptive SPODEs, namely SODE, which uses immunity theory in artificial immune systems to automatically and self-adaptively select the weight for each single SPODE.
- 52
- Hybrid dynamic k-nearest-neighbour and distance and attribute weighted method for classification52
An improved algorithm called dynamic KNN with distance and attribute weighted, simply DKNDAW is singled out and experimentally tested in Weka system, showing that it significantly outperforms other algorithms in terms of the classification accuracy.
- On Fake News Detection with LLM Enhanced Semantics Mining51
A topic model together with a set of specially designed prompts to extract topics and real entities from LLMs and model the relations among news, entities, and topics as a heterogeneous graph to facilitate investigating news semantics is proposed.
- Deep graph level anomaly detection with contrastive learning50
This paper aims to design a new graph level anomaly evaluation way, which first utilizes the contrastive learning strategy to enhance different level representations of normal graphs from node and graph levels by a graph convolution autoencoder with perturbed graph encoder.
- 50
- A Comprehensive Survey on Collaborative Data-access Enablers in the IIoT49
This article surveys the four pillars of information science that enable collaborative data access in an IIoT—standardization, data acquisition, data fusion, and scalable architecture—to provide an up-to-date audit of current research in the field.
- A Destination Prediction Network Based on Spatiotemporal Data for Bike‐Sharing49
An innovative deep learning model to predict the most probable destination for each user in the bike-sharing market, called destination prediction network based on spatiotemporal data (DPNst).
- Collaborative Dynamic Sparse Topic Regression with User Profile Evolution for Item Recommendation49
A novel method CDUE for time-aware item recommendation is presented, which captures the evolution of both user's interests and item's contents information via topic dynamics and adapts a vector autoregressive model to profile users' dynamic interests.
- Towards Graph-level Anomaly Detection via Deep Evolutionary Mapping47
This work proposes a novel deep evolutionary graph mapping framework named GmapAD1, which can adaptively map each graph into a new feature space based on its similarity to a set of representative nodes chosen from the graph set, and can be regarded as a benchmark for explaining anomalies.
- 47
- Hyperbolic Geometric Graph Representation Learning for Hierarchy-imbalance Node Classification46
A novel hyperbolic geometric hierarchy-imbalance learning framework, named HyperIMBA, is proposed to alleviate the hierarchy-IMbalance issue caused by uneven hierarchy-levels and cross-hierarchy connectivity patterns of labeled nodes.
- 46
- Graph Stochastic Neural Networks for Semi-supervised Learning45
A learnable graph neural network coupled with a high-dimensional latent variable to model the distribution of the classification function, and further adopt the amortised variational inference to approximate the intractable joint posterior for missing labels and the latent variable is introduced.
- 45
- 44
- 44
- State of the Art and Potentialities of Graph-level Learning43
The representative algorithms are Frames the representative algorithms into a systematic taxonomy covering traditional learning, graph-level deep neural networks, graph-level graph neural networks, and graph pooling to provide an in-depth analysis.
- 43
- 43
- 43
- Improved detection of landmarks on 3D human face data43
A geometric methodology is developed that can automatically locate 10 established landmark points and 7 other supporting points on human 3D facial scans and show a marked improvement to prior results in the recent literature.
- 42
- Automating DBSCAN via Deep Reinforcement Learning41
A novel Deep Reinforcement Learning guided automatic DBSCAN parameters search framework, namely DRL-DBSCAN, which learns the optimal clustering parameter search policy for different feature distributions via interacting with the clusters, using a weakly-supervised reward training policy network.
- Finding the best not the most: regularized loss minimization subgraph selection for graph classification41
A regularized loss minimization-driven feature selection method for graph classification which integrates subgraph selection and model learning into a unified framework to find discriminative subgraphs with guaranteed minimum loss w.r.t. the objective function is proposed.
- 40
- Trip Reinforcement Recommendation with Graph-based Representation Learning39
A Graph-based Representation Method for Reinforce Trip Recommendation framework for generating a personalized trip with maximal user travel experience and the results demonstrate the superiority of GRM-RTrip compared with the state-of-the-art trip recommendation methods.
- A naive Bayes probability estimation model based on self-adaptive differential evolution39
The experimental results show that the estimation model is efficient and the proposed approach significantly outperforms the traditional probability estimation approaches especially for large data sets (large number of instances and attributes).
- A less-greedy two-term Tsallis Entropy Information Metric approach for decision tree classification38
A less-greedy two-term Tsallis Entropy Information Metric algorithm with a new split criterion and a new construction method of decision trees that yields statistically significantly better decision trees and is more robust to noise.
- Multi-Graph Learning with Positive and Unlabeled Bags38
A new multi-graph learning task with only positive and unlabeled bags, where labels are only available for bags but not for individual graphs inside the bag, is formulated and a puMGL learning framework is proposed which relies on two iteratively combined processes for multigraph learning.
- Heterogeneous Subgraph Transformer for Fake News Detection37
This work proposes a heterogeneous subgraph transformer HeteroSGT to exploit subgraphs in the authors' constructed heterogeneous graph concerning the relations among news topics, entities, and content and reveals that fake news can be effectively detected in terms of the atypical heterogeneous subgraphs centered on them.
- Minimum Entropy Principle Guided Graph Neural Networks37
A novel framework called Minimum Graph Entropy principle-guided Dimension Estimation, i.e. MGEDE, that learns the appropriate embedding dimensions for both node and graph representations is proposed.
- Mining User-aware Multi-relations for Fake News Detection in Large Scale Online Social Networks36
A dual-layer graph is constructed to extract multi-relations of news and users in social networks to derive rich information for detecting fake news and the superiority of Us-DeFake which outperforms all baselines is illustrated.
- eRiskCom: an e-commerce risky community detection platform35
eRiskCom, an e-commerce risky community detection platform to detect risky groups containing identified fraudsters and other closely related users, is designed and tested to demonstrate that it is effective and easy to deploy for real-world applications.
- Reinforcement Learning-Based Dialogue Guided Event Extraction to Exploit Argument Relations35
This paper leverages knowledge of the already extracted arguments of the same sentence to determine the role of arguments that would be difficult to decide individually, and uses the newly obtained information to improve the decisions of previously extracted arguments.
- Detecting and Assessing Anomalous Evolutionary Behaviors of Nodes in Evolving Social Networks35
This study focuses on detecting and assessing AEBN, whose evolutionary mechanisms deviate from their past mechanisms, and proposes a link prediction detection (LPD) method and a matrix perturbation assessment (MPA) method.
- 35
- 35
- 34
- 34
- From Known to Unknown33
A Quality-aware Self-improving Graph Neural Network (QSGNN) which extends the knowledge from known to unknown by leveraging the best of known samples and reliable knowledge transfer and proposes a novel supervised pairwise loss with an additional orthogonal inter-class relation constraint to train the backbone GNN encoder.
- Enhancing Graph Neural Networks by a High-quality Aggregation of Beneficial Information33
This work proposes a flexible method EGAI (Enhancing Graph neural networks by a high-quality Aggregation of beneficial Information), which improves the performance for both shallow and deep GNN models, and to some extent, mitigates over-smoothing.
- 33
- KGTrust: Evaluating Trustworthiness of SIoT via Knowledge Enhanced Graph Neural Networks32
A novel knowledge-enhanced graph neural network (KGTrust) is proposed that utilizes heterogeneous graph structure, node semantics, and augmented trust relationships to learn node embeddings from the perspective of a user as a trustor or a trustee, effectively capturing multi-aspect properties of SIoT trust during information propagation.
- 32
- 32
- 32
- 29
- Uncertainty-Guided Boundary Learning for Imbalanced Social Event Detection29
This work designs a novel uncertainty-guided contrastive learning loss, namely UCL and its variant—UCL-EC, to manipulate distinguishable representation distribution for imbalanced data and combines the results of multi-view evidential classifiers via the Dempster-Shafer theory under the supervision of an additional calibration method.
- 29
- 28
- 28
- Graph Anomaly Detection with Few Labels: A Data-Centric Approach28
A diffusion-based graph generation method to synthesize training nodes, which can be promptly integrated to work with existing anomaly detectors, and proposed two principles for designing denoising neural networks tailored to graph anomaly generation.
- 28
- 27
- 27
- Multi-graph-view subgraph mining for graph classification27
This paper derives an evaluation criterion to estimate the discriminative power and redundancy of subgraph features across all views, with a branch-and-bound algorithm being proposed to prune subgraph search space.
- Opinion Maximization in Social Trust Networks26
This study focuses on social trust networks (STNs), which have the significant characteristics ignored in the previous studies and generalized a famed continuous-valued opinion dynamics model for STNs, which is more consistent with real scenarios.
- 26
- DrugPilot: LLM-based Parameterized Reasoning Agent for Drug Discovery25
DrugPilot is introduced, a LLM-based agent system with a parameterized reasoning architecture designed for end-to-end scientific workflows in drug discovery and significantly outperforms state-of-the-art agents such as ReAct and LoT under the Berkeley function-calling benchmark.
- Heterogeneous Graph Neural Network for Privacy-Preserving Recommendation25
This work designs a two-stage pipeline framework, which includes the privacy-preserving feature encoder and the heterogeneous link reconstructor with gradients perturbation based on differential privacy to tolerate data diversity and against the attack, and designs a bi-level optimization pattern to allocate a suitable privacy budget for the above two modules.
- 25
- Temporal Network Embedding with High-Order Nonlinear Information25
By training the proposed deep guided auto-encoder with a specific mini-batch stochastic gradient descent algorithm, HNIP can efficiently preserves the temporal proximities and highly nonlinear structure of temporal networks.
- 25
- 24
- MeMu: Metric correlation Siamese network and multi-class negative sampling for visual tracking24
A Siamese network-based architecture to embed the metric correlation filter module mentioned above in order to benefit from the powerful representation ability of deep learning and a sampling approach that includes three different kinds of negative samples.
- Self-Representative Manifold Concept Factorization with Adaptive Neighbors for Clustering24
This work proposes a novel manifold concept factorization model with adaptive neighbor structure to learn a better affinity matrix and clustering indicator matrix at the same time and presents superior performance over the state-of-the-art alternatives on numerous datasets.
- Survey and open problems in privacy-preserving knowledge graph: merging, query, representation, completion, and applications23
This paper summarizes the open problems in privacy-preserving KG from four aspects, i.e., merging, query, representation, and completion, and proposes possible technical solutions for them.
- 23
- Adaptive knowledge subgraph ensemble for robust and trustworthy knowledge graph completion23
Experimental results show that the robustness of the ensemble framework outperforms exiting knowledge graph embedding approaches on manually injected noise as well as inherent noisy extracted KGs.
- Calibration error for dual-camera digital image correlation at microscale23
It is concluded that the microscopic dual-camera system is an economic, precise system for 3D profilometry and deformation measurement.
- 22
- Global and Local Influence-based Social Recommendation22
A new global and local influence-based social recommendation model is introduced based on the observation that user purchase behaviour is influenced by both global influential nodes and the local influential nodes of the user, and incorporated into a matrix factorization-based recommendation model.
- Toward value difference metric with attribute weighting22
A new approach to improve VDM by attribute weighting is proposed by incorporating the learned attribute weights into the conditional probability estimates of AWVDM and a modified attribute-weighted value difference metric (MAWVDM).
- Event Extraction by Associating Event Types and Argument Roles21
This model takes into account type and role associations during EE, enabling implicit information sharing among them, and consistently outperforms most state-of-the-art EE methods in both sub-tasks.
- 21
- Curvature Graph Generative Adversarial Networks21
This paper proposed a novel Curvature Graph Generative Adversarial Networks method, named CurvGAN, which is the first GAN-based graph representation method in the Riemannian geometric manifold, and shows superior robustness and generalization.
- Evidential Temporal-aware Graph-based Social Event Detection via Dempster-Shafer Theory21
This work proposes ETGNN, a novel Evidential Temporal-aware Graph Neural Network, which constructs view-specific graphs whose nodes are the texts and edges are determined by several types of shared elements respectively, and introduces a novel temporal-aware aggregator which assigns weights to neighbours according to an adaptive time exponential decay formula.
- Intelligent Data Collaboration in Heterogeneous-device IoT Platforms21
Intelligent data collaboration is presented, i.e., the concept of deep learning providing IoT with the ability to adaptively collaborate to accomplish a task and a mathematical model in general form is presented.
- 21
- Towards large-scale social networks with online diffusion provenance detection21
This work proposes an online regression model for real-time diffusion provenance identification of large networks that first uses offline collected network cascades to infer the edge transmission weights, and then uses an online l 1 non-convex regression model as the identification model.
- 20
- 20
- Keywords-enhanced Deep Reinforcement Learning Model for Travel Recommendation20
This article formalizes travel recommendation as a Markov Decision Process and implement it upon the Actor–Critic framework and integrates keyword information into the reinforcement learning–(RL) based recommendation framework by devising novel state representation and reward function and learns the travel recommendation and keywords generation simultaneously.
- 20
- A fairness-aware multi-stakeholder recommender system20
Experimental results against real-world datasets show that the proposed method significantly improves the diversity of recommended items in a personalized matter and the coverage of providers with no or minor loss of accuracy.
- CenEEGs20
A novel centroid-based EEG selection approach named CenEEGs is proposed, which uses a scale-and-shift-invariance similarity metric to measure similarities of EEG signals and then applies a globally optimal centroid strategy to select valid EEG signals with respect to a similarity threshold.
- Conformal Graph-level Out-of-distribution Detection with Adaptive Data Augmentation19
This work proposes the Conformal Graph-level Out-of-distribution Detection (CGOD), extending the theory of conformal prediction to graph-level OOD detection with a rigorous control over the false positive rate and develops a new aggregated non-conformity score function based on the proposed adaptive data augmentation.
- 18
- Cross-Domain Facial Expression Recognition via Disentangling Identity Representation18
This work exploits the idea of domain generalization (DG) and proposes a representation disentanglement model that can disentangle expression features from expression-irrelevant ones and exhibit sufficient domain invariance and discriminative ability.
- Robust face alignment by dual-attentional spatial-aware capsule networks18
This work proposes a face alignment algorithm named Dual-attentional Spatial-aware Capsule Network (DSCN), which is more robust to occlusion and outperforms state-of-the-art methods in the literature.
- 18
- Deep Multi-Attributed-View Graph Representation Learning17
A novel deep unsupervised graph representation learning model, called Multi-attributed-view graph Convolutional AutoEncoder (MagCAE), which learns the node-pairwise proximity among multi-attribution views and node embeddings, across which a novel loss function is designed to preserve the nodes- Pairwise likelihood.
- 17
- Cross-Network Social User Embedding with Hybrid Differential Privacy Guarantees17
A Cross-network Social User Embedding framework, namely DP-CroSUE, is proposed to learn the comprehensive representations of users in a privacy-preserving way and a novel cross-network GCN embedding model is designed to transfer knowledge across networks through those aligned users.
- 17
- 16
- 16
- Boosting for graph classification with universum16
A new universum graph classification framework which leverages additional “non-example” graphs to help improve the graph classification accuracy and proposes a mathematical programming algorithm, ugBoost, which integrates discriminative subgraph selection and margin maximization into a unified framework to fully exploit the universum.
- Automated face extraction and normalization of 3D Mesh Data16
The goal of this paper is to develop a reliable automatic methodology for extracting the face from raw data with texture acquired from a stereo imaging system, based on the medical researchers' specific requirements.
- 15
- 15
- Intra- and inter-association attention network-enhanced policy learning for social group recommendation15
This work proposes an Intra- and inter-association attention network with Policy learning for Social Group Recommendation (IP-SGR), and proposes a reinforcement learning-based model to obtain the positive and negative reward for social group recommendation.
- Fair-SRS: A Fair Session-based Recommendation System15
This paper demonstrates Fair-SRS, a Fair Session-based Recommendation System that predicts user's next click based on their historical and current sessions, and optimizes recommendations by making a trade-off between accuracy and personalized diversity.
- 15
- Parallel DNN Inference Framework Leveraging a Compact RISC-V ISA-based Multi-core System15
This paper proposes a collaborative RISC-V multi-core system for Deep Neural Network (DNN) accelerators and reveals that a multi- core design along with an appropriate scheduling algorithm can efficiently decrease the runtime and elevate the throughput.
- TLR: Transfer Latent Representation for Unsupervised Domain Adaptation15
This manuscript designs an objective function based on a simple linear autoencoder to derive the latent representations of both domains, and develops a novel method, transfer latent representation (TLR), to learn a better latent space.
- 15
- 15
- Exploring the occupational biases and stereotypes of Chinese large language models14
This study systematically examines biases in five representative C-LLMs to contribute substantially to the literature on biases in C-LLMs and provide important insights for users aiming to utilize these models more effectively and ethically.
- Causal Variational Inference for Deconfounded Multi-Behavior Recommendation14
Causal Variational Inference for Deconfounded (CVID) MBR employs a variational graph autoencoder to model latent uncertainty in multi-behavior interactions and introduces a confounder inference module to generate behavior-specific latent confounders via variational inference.
- Global Interpretable Graph-level Anomaly Detection via Prototype14
This work proposes a global interpretable Graph-Level Anomaly Detection model via Prototype (GLADPro), which provides global-level explanations throughout the entire dataset, that is, the significant subgraph patterns that consistently influence the model's decisions.
- Heterogeneous Graph Neural Network via Knowledge Relations for Fake News Detection14
This work proposes a novel Heterogeneous Graph Neural Network via Knowledge Relations for Fake News Detection (HGNNR4FD), and empirically demonstrates that the framework achieves superior results and yields improvement over the baselines regarding evaluation metrics of accuracy, precision, recall, and F1-score on four real-world datasets.
- 14
- 14
- 13
- CloudDenseNet: Lightweight Ground-Based Cloud Classification Method for Large-Scale Datasets Based on Reconstructed DenseNet13
The lightweight CloudDenseNet structure is designed meticulously according to the distinctive characteristics of ground-based clouds and the intricacies of large-scale diverse datasets, which amplifies the generalization ability and elevates the recognition accuracy of the network.
- Dual-branch Density Ratio Estimation for Signed Network Embedding13
Extensive empirical results demonstrate that DDRE not only significantly outperforms the methods based on density ratio estimation but also achieves competitive performance compared with other types of methods such as graph likelihood, generative adversarial networks, and graph convolutional networks.
- 13
- 13
- VCF: Velocity Correlation Filter, Towards Space-Borne Satellite Video Tracking13
Experimental results show the proposed approach outperforms state-of-the-art tracking methods with a more than 100 frames per second, employing velocity feature and inertia mechanism to construct a kernel correlation filter for satellite video targets tracking.
- DAliM: Machine Learning Based Intelligent Lucky Money Determination for Large-Scale E-Commerce Businesses13
A machine-learning based lucky money determination approach, called DAliM, is proposed for e-commerce businesses to achieve their desired goals and demonstrates that it manages to decrease the lucky money spent and increase the final purchase rate compared to the state-of-the-art baseline.
- Combining Structured Node Content and Topology Information for Networked Graph Clustering13
This article forms a new networked graph clustering task where a network contains a set of inter-connected super-nodes, each of which is a single-attribute graph, and proposes an Attributed Random Walk Kernel (ARWK) to calculate the similarity between super-Nodes.
- Dual influence embedded social recommendation13
This paper first simulates the social influence diffusion in the network to find the global and local influence nodes and then embeds this dual influence data into a traditional recommendation model to improve accuracy.
- 13
- Multi-graph-view learning for complicated object classification13
A multi-graph-view bag learning algorithm (MGVBL), which aims to explore subgraph features from multiple graphviews for learning by enabling a joint regularization across multiple graph-views, and enforcing labeling constraints at the bag and graph levels, is able to discover most effective sub graph features across all graph- views for learning.
- Multi-Instance Learning from Positive and Unlabeled Bags13
The main challenge of puMIL is to accurately identify negative bags for training discriminative classification models, and an Artificial Immune System based self-adaptive process to select most reliable negative bags in each iteration is solved.
- 12
- 12
- Toward Cross-Lingual Social Event Detection with Hybrid Knowledge Distillation12
A GNN-based framework that integrates cross-lingual word embeddings into the process of graph knowledge distillation for detecting events in low-resource language data streams and is highly effective at detection in languages where training samples are scarce.
- An Efficient Automatic Meta-Path Selection for Social Event Detection via Hyperbolic Space12
A Heterogeneous Information Graph representation via Hyperbolic space combined with an Automatic Meta-path selection (GraphHAM) model, an efficient framework that automatically selects the meta-path's weight and combines hyperbolic space to learn information on social media.
- 12
- An Interpretable Brain Graph Contrastive Learning Framework for Brain Disorder Analysis12
An interpretable brain graph contrastive learning framework, which aims to learn brain graph representations by a unsupervised way for disorder prediction and pathogenic analysis, which can observe salient brain regions and connections for pathogenic analysis.
- Contrastive Learning Drug Response Models from Natural Language Supervision12
The CLDR converts regression labels into text, which is merged with the drug response caption as a second sample modality instead of the traditional modes, i.e., graphs and sequences to improve the continuous text representation.
- 12
- 12
- GARAT: Generative Adversarial Learning for Robust and Accurate Tracking12
This work proposed a novel and concise generative adversarial learning method to solve the problem especially when the target is going under drastic changes of appearance, illumination variations and background clutters.
- 12
- 12
- 12
- Nonlocal Patch Based t-SVD for Image Inpainting: Algorithm and Error Analysis12
A novel image inpainting framework consisting of an interpolation step and a low-rank tensor completion step that can be decomposed into two components, one of which can be bounded by a reasonable assumption named local patch similarity, and another part is lower than that using matrix.
- Link sign prediction by Variational Bayesian Probabilistic Matrix Factorization with Student-t Prior12
A novel Variational Bayesian Probabilistic Matrix Factorization with Student-t prior model (TBMF) is proposed, which improves the prediction performance significantly compared with the state-of-the-art algorithms, especially when the observed links are few.
- 12
- Shape-Based Classification of 3D Head Data12
Several different craniofacial descriptors that are being used in research studies for two cranioFacial disorders: the 22q11.2 deletion syndrome and plagiocephaly/brachycephania, disorders caused by pressure on the head are introduced.
- 11
- Heterogeneous Social Event Detection via Hyperbolic Graph Representations11
This study proposes two hyperbolic graph representation-based methods for detecting social events from heterogeneous social media environments and designs a HSED model that converts complex social information into a unified social message graph.
- 11
- Functional connectivity learning via Siamese-based SPD matrix representation of brain imaging data11
Experimental results indicate that SiameseSPD-MR can capture more significant differences in functional connectivity between neural states than the state-of-the-art counterparts do, and the obtained functional connectivity representations conforming to the proposed measure can act as meaningful markers for brain network analysis and ASD discrimination.
- CloudDeepLabV3+: a lightweight ground-based cloud segmentation method based on multi-scale feature aggregation and multi-level attention feature enhancement11
A lightweight ground-based cloud image adaptive segmentation method named CloudDeepLabV3+ that integrates multi-scale features aggregation and multi-level attention feature enhancement that promotes the accuracy of cloud image feature extraction while reducing the loss of detailed features is designed.
- Don't Ignore Alienation and Marginalization: Correlating Fraud Detection11
COFRAUD is proposed, a correlation-aware fraud detection model, which innovatively incorporates synergistic camouflage into fraud detection and captures the correlation among multi-relation fraudulent behaviors.
- 11
- 11
- 11
- 11
- Let’s CoRank: trust of users and tweets on social networks11
A Coupled Dual Networks Trust Ranking (CoRank) method to evaluate the trustworthiness of users and tweets by analysing user/tweet behaviours on Twitter is developed and compared with three baseline methods PageRank, TURank, and Weighted PageRank to show how the approach outperforms the existing ones.
- Exploring Features for Complicated Objects: Cross-View Feature Selection for Multi-Instance Learning11
A new cross-view feature selection problem is formulated that aims to identify the most representative features across all feature views for MIL, and the solution to the objective function will ensure that the identified top-m features are the most informative ones across allfeature views.
- Learning to Rank the Severity of Unrepaired Cleft Lip Nasal Deformity on 3D Mesh Data11
A grid-patch based measurement of symmetry is introduced, with which a computer program learns to rank the severity of cleft lip on 3D meshes of human infant faces and shows that the rankings predicted by the proposed features were highly correlated with the ranking orders provided by experts that were used as the ground truth.
- 11
- Learning to compute the plane of symmetry for human faces11
This paper presents a method to compute the plane of symmetry for 3D meshes of the human head and face through learning that performs better than the existing mirror method, and is robust to rotation and noise.
- 10
- 10
- 10
- 10
- Unbiased and Efficient Self-Supervised Incremental Contrastive Learning10
A self-supervised Incremental Contrastive Learning framework consisting of a novel Incremental InfoNCE (NCE-II) loss function by estimating the change of noise distribution for old data to guarantee no bias with respect to the retraining, and a meta-optimization with deep reinforced Learning Rate Learning (LRL) mechanism which can adaptively learn the learning rate according to the status of the training processes and achieve fast convergence which is critical for incremental learning.
- Community Preserving Social Recommendation with Cyclic Transfer Learning10
This work proposes a community-preserving recommendation framework with cyclic transfer learning, incorporating heterogeneous community influence into the rating domain, and analyzes the characteristics of the community domain and its inter-influence on the ratingdomain, and construct link constraints and preference constraints in the communitydomain.
- SALE: Self-adaptive LSH encoding for multi-instance learning10
A self-adaptive LSH encoding method for MIL, termed as SALE, which efficiently deals with large MIL problems, due to its low complexity and RSH’s ability to exploit key information of MIL.
- 9
- Zero-shot Learning for Preclinical Drug Screening9
A zero-shot learning solution for the DRP task in preclinical drug screening, called MSDA, which can be seamlessly integrated with conventional DRP methods, learning invariant features from the prior response data of similar drugs to enhance real-time predictions of unlabeled compounds.
- 9
- Graph structure reforming framework enhanced by commute time distance for graph classification9
This work proposes a plug-and-play framework based on Commute Time Distance (CTD), in which information is propagated in commute time distance neighborhoods and can capture higher-order structural information by utilizing commute paths to enhance the expressive power of GNNs.
- 9
- 9
- Knowledge-aware document summarization: A survey of knowledge, embedding methods and architectures9
This paper pursues to present the first systematic survey for the state-of-the-art methodologies that embed knowledge into document summarizers, and proposes novel taxonomies to recapitulate knowledge and knowledge embeddings under the document summarization view.
- Glider: rethinking congestion control with deep reinforcement learning9
An extensive array of experiments on Pantheon is used to show that Glider can adapt well to varying buffer sizes and is resilient to random loss, and comparable performance as other learning-based congestion control protocols in the literature.
- 9
- 9
- Brain EEG Time Series Selection: A Novel Graph-Based Approach for Classification9
This paper proposes a novel maximum weight clique-based EEG selection approach, named mwcEEGs, to map EEG selection to searching maximum similarity-weighted cliques from an improved Fr\'{e}chet distance-weighting undirected EEG graph simultaneously considering edge weights and vertex weights.
- Bernoulli random forests: closing the gap between theoretical consistency and empirical soundness9
A Bernoulli random forests model (BRF), which intends to close the gap between the theoretical consistency and the empirical soundness of random forests classification and demonstrates the best performance among all theoretical random forests.
- 9
- Mining Frequent Subgraph by Incidence Matrix Normalization9
This paper proposes a fast algorithm for mining frequent subgraphs in large database of labeled graphs that uses incidence matrix to represent the labeled graphs and to detect their isomorphism.
- 8
- 8
- 8
- TigeCMN: On exploration of temporal interaction graph embedding via Coupled Memory Neural Networks8
A novel framework named TigeCMN is proposed to learn node representations from a sequence of temporal interactions and can achieve significant gains compared with recent state-of-the-art baselines, and devise two coupled memory networks to store and update node embeddings in the external matrices explicitly and dynamically.
- 8
- On the Minimum Differentially Resolving Set Problem for Diffusion Source Inference in Networks8
By the weighted set cover theory, an approximation algorithm with the Θ(ln n) approximability for the minimum DRS problem on general graphs, where n is the graph size is proposed.
- 8
- An Empirical Study on Several Classification Algorithms and Their Improvements8
The paper studies and compares the classification performance on classification accuracy empirically, using the whole 36 UCI data sets obtained from various sources selected by Weka, and re-demonstrate the efficiency of all these improved algorithms.
- 7
- Text is All You Need: LLM-enhanced Incremental Social Event Detection7
A novel framework, LLM-enhanced Social Event Detection (LSED) is proposed, which leverages the rich background knowledge of LLMs to address this challenge, and introduces hyperbolic space embeddings, which are more suitable for natural language sentence representations, to enhance clustering performance.
- 7
- Graph-Structured Small Molecule Drug Discovery Through Deep Learning: Progress, Challenges, and Opportunities7
This paper provides an overview of the major tasks in small-molecule drug discovery and their interrelationships, and analyzes the related methods, commonly used datasets, and technological development trends.
- 7
- Towards a better negative sampling strategy for dynamic graphs7
This work introduces an innovative negative sampling approach named Enhanced Negative Sampling (ENS), which takes into account two pervasive traits observed in dynamic graphs, and employs a designed scheduling function to strategically control the progression of difficulty of the negative samples throughout the training.
- 7
- 7
- 7
- 7
- 7
- Robust correlation filter tracking with multi-scale spatial view7
It is confirmed by the experiments and analysis that the RCFMSV tracking method proposed in this work is competitive with the state-of-the-art methods in tracking performance.
- Noise-Resilient Similarity Preserving Network Embedding for Social Networks7
This paper aims to exploit node similarity to address the problem of social network embedding with noise and proposes a node similarity preserving (NSP) embedding method that exploits a comprehensive similarity index to quantify the authenticity of the observed network structure.
- 7
- 7
- 7
- 7
- A Combined Classification Algorithm Based on C4.5 and NB7
This paper presents a combined classification algorithm based on C4.5 and NB, simply C 4.5-NB, which significantly outperforms C3.7 and NB in terms of classification accuracy and AUC.
- 6
- 6
- Generative Models for Time Series Anomaly Detection: A Survey6
A range of generative models are investigated, offering mathematical summaries of the predominant techniques in TSAD and recommendations for appropriate generative methods tailored to various application domains are proposed.
- 6
- 6
- Adaptive curvature exploration geometric graph neural network6
The Adaptive Curvature Exploration Geometric Graph Neural Network is proposed to automatically learn high-quality graph representations and explore the embedding space with optimal curvature at the same time to optimize the multi-objective optimization problem of the graph representation learning and curvature exploration.
- ReInCre: Enhancing Collaborative Filtering Recommendations by Incorporating User Rating Credibility6
ReInCre is the first work to incorporate the rating credibility of users in a CF recommendation, and works as a powerful add-on to existing CF-based recommender systems in order to optimize the neighborhood.
- 6
- 6
- Hi-GMAE: Hierarchical Graph Masked Autoencoders5
Hierarchical Graph Masked AutoEncoders (Hi-GMAE), a novel multi-scale GMAE framework designed to handle the hierarchical structures within graphs, which consistently demonstrates that Hi-GMAE outperforms 29 state-of-the-art self-supervised competitors in capturing comprehensive graph information.
- 5
- Adaptive User Dynamic Interest Guidance for Generative Sequential Recommendation5
A novel generative sequential recommendation framework named ADIGRec (Adaptive User Dynamic Interest Guidance for Generative Sequential Recommendation), which adaptively focuses on users' dynamic interest features and inherent interest features encoded from historical sequences as new guidance conditions is proposed.
- 5
- 5
- 5
- 5
- 5
- 5
- 5
- 5
- Modeling the social influence of COVID-19 via personalized propagation with deep learning5
This work extended DeepInf such that it can predict the social influence of COVID-19 via the transition probability of the page rank domain and gives rise to a deep learning-based personalized propagation algorithm, called DeepPP, which provides more accurate social influence predictions.
- 5
- Ensemble Learning on Hyperspectral Remote Sensing Image Classification5
This paper takes the features of hyperspectral remote sensing data and classification algorithms as the background, applying the ensemble learning to image classification, and shows that ensemble learning on hyperspectrals can achieve higher classification accuracy.
- Using Wittgenstein’s Family Resemblance Principle to Learn Exemplars5
The paper presents a model capable of learning exemplars using the principle of family resemblance and adopting Bayesian networks for a representation of exemplars and shows promising results that suggest that previous assumptions about the way the authors categories need reopening.
- Enhanced Social Event Detection through Dynamically Weighted Meta-Paths Modeling4
The Dynamically Weighted Meta-Paths Modeling (DWMM) framework for large-scale social event detection is presented and extensive experiments show that DWMM enhances performance, demonstrating its effectiveness.
- 4
- 4
- 4
- Self-supervised Time-aware Heterogeneous Hypergraph Learning for Dynamic Graph-level Classification4
DyH2GNet is proposed, a novel time-aware self-supervised heterogeneous hypergraph neural network that models complex, non-pairwise interactions and their temporal dynamics by integrating k-hop nodes and attribute correlations into a heterogeneous hypergraph.
- Complex query answering over knowledge graphs foundation model using region embeddings on a lie group4
A Lie group is introduced as a compact embedding space for complex query embedding, enhancing ability to handle the intricacies of knowledge graphs the foundation model and complex query processing.
- Sea ice detection network for icebreakers in polar environments with attention-based deeplabv3+ architecture4
A sea ice detection method that combines multi-branch attention feature alignment and multi-scale feature extraction is proposed, which strengthens the model’s extraction of sea ice features and elevates representation performance.
- VR-GNN: variational relation vector graph neural network for modeling homophily and heterophily4
This work proposes a novel GNN model based on relation vector translation named as Variational Relation Vector Graph Neural Network (VR-GNN), which gains consistent and significant improvements against existing strong GNN methods under heterophily and competitive performance under homophily.
- Graph explicit pooling for graph-level representation learning4
A novel Graph explicit Pooling (GrePool) method is introduced, which selects nodes by explicitly leveraging the relationships between the nodes and final representation vectors crucial for classification, which outperforms 14 baseline methods for most datasets and enhances GrePool's performance without incurring additional computational costs.
- 4
- IEEE Access Special Section Editorial: Advanced Data Mining Methods for Social Computing4
This IEEE ACCESS Special Section on advanced data mining methods for social computing aims at bringing together researchers to disseminate their findings in the quickly growing field of social computing.
- 4
- 4
- GroExpert: A Novel Group-Aware Experts Identification Approach in Crowdsourcing4
The proposed GroExpert approach employs a fully connected neural network to implement the pairwise ranking method when identifying experts based on their relative values in their working groups, which can be used as a basis for quality estimation in crowdsourcing.
- CoTrRank: Trust Evaluation of Users and Tweets4
This approach models users and tweets in two coupled networks and calculate their trust values in different trust spaces and provides a configurable way when mapping the calculated raw evidences to the trust values.
- Multiple-Instance Learning with Evolutionary Instance Selection4
A novel learning method, MILEIS (Multiple-Instance Learning with Evolutionary Instance Selection), to adaptively determine the informative instances for feature mapping that can adjust itself to the data without explicit specification of functional or distributional form for the underlying model.
- Semi-data-driven network coarsening4
This paper presents a new semi-data-driven network coarsening model to learn coarsened networks by embedding both static network structure data and dynamic network information spreading data and proves that the learning model is convex and the Accelerated Proximal Gradient algorithm is adapted to achieve the global optima.
- Locally Weighted Learning: How and When Does it Work in Bayesian Networks?4
Experiments and comparisons demonstrate that locally weighting technologies just slightly outperforms unweighted complex BNCs on ACC and AUC, which means that although locally weighting could significantly improve the performance of NB, it could not work well on BNCs with complex structures.
- Craniofacial Image Analysis4
A craniofacial image analysis pipeline is described and the computational methods developed by the Multimedia Group at the University of Washington including data acquisition and preprocessing, low- and mid-level features, quantification, classification, and content-based retrieval are introduced.
- Learning the attribute selection measures for decision tree4
The paper considers a number of different attribute selection measures and experimentally examines their behavior in classification, and shows that the choice of measure doesn’t affect the classification accuracy, but the size of the tree is influenced significantly.
- Attribute Weighting via Differential Evolution for Attribute Weighted Clonal Selection Algorithm4
This paper uses differential evolution (DE) algorithm to determine the weights of attributes and then uses these weights in AWCSA, and evaluates the performance of new algorithm (DE-AWCSA) on six standard datasets.
- 4
- 4
- 4
- 4
- STGS: Spatio-temporal Graph Sparsification Using Reinforcement Learning3
STGS (Spatio-Temporal Graph Sparsification), a reinforcement learning-based framework for sparsifying spatio-temporal graphs, is introduced, demonstrating that STGS outperforms existing methods in both structural preservation and downstream forecasting tasks.
- CFI-Former: Efficient lane detection by multi-granularity perceptual query attention transformer3
A multi-granularity perceptual query attention transformer lane detection method, CFI-Former, is proposed to achieve more accurate lane detection and a weighted adaptive LIoU loss (Lφ-LIoU) is proposed to improve lane detection in more challenging scenarios by adaptively increasing the relative gradient of high IoU lane objects and the weight of the loss.
- 3
- 3
- 3
- Heterogeneous graph neural network with semantic-aware differential privacy guarantees3
A novel Hete rogeneous graph neural network with HeteSDG, which provides a double privacy guarantee and performance trade-off in terms of both graph features and topology, and designs a two-stage mechanism, which includes the feature attention personalized mechanism and the topology gradient perturbation mechanism.
- 3
- 3
- 3
- 3
- 3
- Research on Risk Prevention Measures of Network Information Security in Power Systems3
The network information security of power systems is studied, the risks faced by power system network operation are analyzed, and some preventive measures are put forward.
- Microcluster-Based Incremental Ensemble Learning for Noisy, Nonstationary Data Streams3
A new incremental ensemble model is presented for classifying nonstationary data streams with noise that integrates three strategies: incremental learning to monitor and adapt to concept drift; ensemble learning to improve model stability; and a microclustering procedure that distinguishes drift from noise and predicts the labels of incoming instances via majority vote.
- Improved Coupled Tensor Factorization with Its Applications in Health Data Analysis3
This paper proposes two improved coupled tensor factorization methods: one for approximately coupled datasets and the other for partially coupled datasets, and demonstrates the improved accuracy of these approaches over existing baselines.
- 3
- A Brain Network Inspired Algorithm: Pre-trained Extreme Learning Machine3
Experiments and comparisons on face image recognition and handwritten image annotation applications demonstrate that P-ELM is promising and achieves superior results compared to the original ELM algorithm and other ELM-based algorithms.
- Graph-theoretic node importance mining in world city networks: methods and applications3
This is the first systematic literature review of node importance from the view of graph-theoretic mining, which focuses on power-embedded fusion measurements in the transmission mechanism-based complex applications.
- Learning attribute weighted AODE for ROC area ranking3
This paper investigates how to learn an AWAODE with accurate ranking from data sets and proposes a new approach to weight AODE for generating accurate ranking, called decision tree-based attribute weighted averaged one-dependence estimator, simply DTWAODE.
- Active AODE learning based on a novel sampling strategy and its application3
Experimental results show that the AODE classification model and the novel active learning strategy can get better classification accuracy with fewer labelled samples than that of the state-of-the-art approaches for active learning.
- Weighted Clone Selection Algorithm based on Rough Set Theory3
In weighted clone selection algorithm attribute weights obtained directly from the training data using rough set theory, the attribute weights was used to test Data classification and verify the validity of the method by the experiments of UCI data sets.
- Omnidirectional Motion Control for the Humanoid Soccer Robot3
A detailed description of the architecture in basic motion using Nao model is given and the team G-Star has worked out the quantitative relation of joint angle in motion control and developed a toolkit to calculate the angle of joints accurately.
- 2
- 2
- 2
- 2
- A Multiagent-based Framework on RoboCup Simulation System for Enhancing Rescue Operation during Dynamic Disaster Environments2
This work proposes a novel framework called adaptive A* graph routing with smoothed obstacle clearance that optimizes route selection and obstacle clearance for police, improving rescue efficiency and achieves superior rescue outcomes compared to the current baselines.
- Domain-Level Disentanglement Framework Based on Information Enhancement for Cross-Domain Cold-Start Recommendation2
A generalized framework that Domain-level Disentanglement framework based on information enhancement for Cross-domain Cold-start Recommendation, which achieves deep utilization of domain-level information through independent extraction of domain knowledge and fusion using heuristic strategies.
- 2
- 2
- 2
- 2
- 2
- GCINet: global convolution interaction network with a pre-trained reversible normalization method for long-term time series forecasting2
A new time series forecasting model, GCINet, is developed, which introduces a new mode of interaction based on a convolutional network that captures both long-term and short-term temporal features while reducing complexity during feature extraction.
- 2
- 2
- 2
- Signed network representation with novel node proximity evaluation2
Three new node proximity metrics were proposed in this study, based on the rigorous theoretical investigation on a new distance metric - signed average first-passage time (SAFT).
- 2
- Real-time monitoring and reminding of remote peritoneal dialysis system based on the principle of least squares2
The remote diagnosis and treatment system of peritoneal dialysis developed by Shenzhen Traditional Chinese Medicine Hospital can monitor the whole course ofPeritoneal Dialysis data of patients and inform the patients of the results according to the confidence interval of the regression prediction, which greatly strengthens the interaction of the system and increases the patients' compliance.
- 2
- 2
- 2
- 2
- Attribute Segmentation for Transformer Substation Using Convolutional Network2
This paper focuses on the pixelwise attribute classification based on convolutional network, and trains the attribute classification network, followed by attribute-specific segmentation model, which achieves a good performance in a new dataset of transformer substation.
- Boosting for superparent-one-dependence estimators2
This paper investigates a novel approach to ensemble the single SPODE based on the boosting strategy, Boosting for superparent-one-dependence estimators, simply, BODE, which first endows every instance a weight, and then finds an optimal SPODE with highest accuracy in each iteration as a weak classifier.
- 2
- Revisiting Graph-Level Anomaly Detection: From Partially to Fully Unsupervised Learning1
SUL is an end-to-end framework that iteratively optimizes anomaly detection and uncertainty modeling via an Expectation-Maximization algorithm and GUL is an end-to-end framework that iteratively optimizes anomaly detection and uncertainty modeling via an Expectation-Maximization algorithm.
- Beyond Smoothing: A Discriminative Enhanced Aggregation Graph Neural Network for Camouflaged Fraud Detection1
This work proposes a novel Discriminative Enhanced Aggregation Graph Neural Network-based FraudDEtectioNMoDel (DEFEND), which incorporates tailored discriminative mechanisms that strengthen representation learning at two complementary levels: intra-relation and inter-relation.
- LHG: LLM-enhanced and Heterogeneous Graph-induced for Unsupervised Social Event Detection1
An unsupervised SED framework with an LLM enhancement and Heterogeneous Graph induction (LHG) is proposed, which addresses the label scarcity problem and achieves state-of-the-art (SOTA) results on two widely used real-world datasets.
- 1
- 1
- Differentially Private Graph Learning: A Survey1
This survey presents the first comprehensive and systematic review of Differentially Private Graph Learning (DPGL), and organizes existing DPGL methods into four categories based on the granularity of privacy protection, namely node-level, edge-level, graph-level, and local differential privacy.
- 1
- Multi-view contrastive learning with Static attributes and Dynamic interests for Sequential Recommendation1
By leveraging contrastive learning, SDSR aligns representations across multiple data views—such as interaction graphs, knowledge graphs, and sequential data—creating a unified user-item model that bridges long-term preferences with short-term trends, and ensures consistency across various representations.
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- Careful Selection and Thoughtful Discarding: Graph Explicit Pooling Utilizing Discarded Nodes1
A novel Graph Explicit Pooling (GrePool) method is introduced, which selects nodes by explicitly leveraging the relationships between the nodes and final representation vectors crucial for classification, and enhances GrePool's performance without incurring additional computational costs.
- Introduction to the Special Issue on Advanced Graph Mining on the Web: Theory, Algorithms, and Applications: Part 11
This special issue is to provide a forum for researchers and practitioners to present their latest research findings and engineering experiences in the theoretical foundations, empirical studies, and novel applications of Graph Mining.
- 1
- VR-GNN: Variational Relation Vector Graph Neural Network for Modeling both Homophily and Heterophily1
A novel GNN model based on relation vector translation named Variational Relation Vector Graph Neural Network (VR-GNN) is proposed, which gains consistent and significant improvements against state-of-the-art GNN methods under heterophily, and competitive performance under homophily.
- Graph-based Semi-Supervised Learning by Strengthening Local Label Consistency1
A novel heuristic pre-processing technique, namelyLocal Label Consistency Strengthening (ŁLCS), which automatically expands new nodes and edges to refine the label consistency within a dense subgraph and couple it with the popular graph convolution network and graph attention network.
- 1
- The Concept Layer, Feature Layer and Landing Layer of Ubiquitous Internet of Things in Power System1
This paper considers the above three dimensions of ubiquitous IOTIPS: what is its implication, what are its inherent attributes and how to get results in information system, digital economy and new business model.
- dFDA-VeD: A Dynamic Future Demand Aware Vehicle Dispatching System1
This paper proposes a dynamic future demand aware vehicle dispatching system that can dynamically search the relocation centers considering both travel demand and traffic conditions and demonstrates that it significantly improves the serving ratio and with a very small increase in operation cost.
- 1
- IEEE Access Special Section Editorial: Advanced Data Analytics for Large-Scale Complex Data Environments1
Novel data analytics methods are needed to discover meaningful knowledge in advanced applications from objects with large-scale characteristics, such as chemical pounds in bio-pharmacy, brain regions in brain networks and users in social networks.
- 1
- 1
- Direct Discriminative Bag Mapping for Multi-Instance Learning1
This paper proposes a direct discriminative mapping approach for multi-instance learning (MILDM), which identifies instances to directly distinguish bags in the new mapping space.
- 1
- 1
- 1
- [A novel SMOreg algorithm based on instance cloned and its research on spectral modeling for hydromica in uranium deposit].1
The authors' adopt SMOreg in the spectral modelling for hydromica, and validate its effectiveness, and propose a novel method called ICSMOreg, which improves the accuracy of prediction, and also reduces the negative impact of noise.
- 1
- A fast algorithm for mining frequent subgraphs1
An algorithm for mining frequent subgraphs based on associated matrix based on the depth-first searching method has higher speed and better stability than other similar ones.
- 1
- –
- Is the Last Check-In All You Need? Next POI Recommendation: Recall and Rerank–
K1-POI is proposed, a concise recall-and-rerank framework that formulates sparse next-POI prediction as candidate generation and calibration and an efficient Spatio-Temporal Prior Reranker (ST-Reranker) that calibrates the ordering within the top-K candidates using additive, model-agnostic spatio-temporal priors computed from the training split.
- Modeling Endogenous Logic: Causal Neuro-Symbolic Reasoning Model for Explainable Multi-Behavior Recommendation–
This work proposes a novel Causal Neuro-Symbolic Reasoning model for Explainable Multi-Behavior Recommendation (CNRE), which operationalizes the endogenous logic by simulating a human-like decision-making process and generates an explainable causal mediator that approximates an ideal state isolated from confounding effects.
- MARCH: Multi-Teacher Contrastive Hypergraph Distillation–
The proposed MARCH proposes a multi-teacher distillation strategy that adaptively transfers complementary knowledge from multiple teachers at both node and hyperedge levels, empowering the student model to learn richer and more discriminative representations and even outperform its teachers.
- –
- –
- The Critical Role of Presettable and Expandable Artificial Surgical Valves in the Lifetime Management of Valvular Heart Disease–
Animal studies and preliminary clinical trials have demonstrated promising outcomes, suggesting these innovations will profoundly impact the sequential management of valvular heart disease under a lifetime management paradigm.
- –
- A Relation-Constraint Link Prediction Model for Dynamic Knowledge Graphs with Entity Drift–
A novel link prediction model based on the embedding representation to handle the incompleteness of KGs with entity drift, termed as DCEL is proposed, which reveals the superiority of DCEL model compared to its SOTAs.
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- Efficient Federated Learning for Feature Aggregation with Heterogenous Edge Devices–
This work forms an optimization problem which minimizes the energy consumption of all devices, under the constraint of model quality, and designs a polynomial-time algorithm for the problem.
- –
- –
- –
- Research on the Topology of the Ubiquitous IoT Distribution Network in Power System–
This paper focuses on the application scenario of distribution network and analyses its network topology and the application prospect of the topology of the proposed ubiquitous IoT distribution network in power system is prospected.
- Analysis of the Information Security Risk Assessment Mechanism Based on the Power System–
The analytic hierarchy process (AHP) is established, a new information security risk assessment model in combination with fuzzy mathematics in risk assessment is designed, and the content of the model is described in detail.
- –
- –
- Attribute Classification for Transformer Substation Based on Deep Convolutional Network–
A new dataset of transformer substation is built for attribute classification, and the deep convolutional network is employed, which outputs the posterior probability of each attribute.
- –
- –
- Mining the crime survey to support crime profiling–
The preprocessing activities that were performed on survey data to allow their use with data mining models and the results of early analysis of the survey data using decision trees and the users' interpretation of these results are reported.
- Research on real-time data processing technology for Internet of things–
A distributed storage and computing architecture for data partition and multi node distribution is proposed that can achieve linear expansion of the processing performance and storage size of the real-time data.
- Multi-graph learning–
This thesis formulates a new multi-graph learning paradigm for representing and classifying complicated objects, where each object to be classified is represented graphs under multi- graph-view.
- Mining Top-k Minimal Redundancy Frequent Patterns over Uncertain Databases–
This paper presents a new algorithm, UTFP, to mine top-k minimal redundancy frequent patterns of length no less than minimum length \(min\_l\) without setting threshold, and proposes a set of strategies to prune and reduce search space.
- A realization of load balancing based on distributed real-time database isomerism cluster–
This paper proposed a realization of load balancing that adopts dynamic feedback mechanism to optimize the communication mechanisms of load information collection, and gives an intelligent allocation method to improve the overall system performance.
- –
- –
- –
- Agent Simulation Group on the Robocup 3D Realization of Basic Motions–
Experiments proved that method of rotation matrix and the ZMP in agent movement calculation meet the accuracy requirements such as real-time and software of dynamic simulation and Evolutionary Algorithm perform well in agent motion analysis and parameter optimization.
- Improved KNN-SMOreg algorithm and its application in predicting the amount of hematite from uranium–
A novel method called WKNN-SMOreg is proposed, which weights the features according to the association of their attributes on the hybrid of KNN and SMOreg, which improves the accuracy of prediction, and reduces the negative impact of the noise.
- –
- –
- –
- –
- –
- –
- FUZZY PROBABILITY SPACES CHARCTERIZED BY FUZZY LATTICES–
The set features of the probability of the fuzzy probability spaces and the correlative theory of fuzzy lattices are obtained and the probability is decompose into the product of a fuzzy lattice homomorphism and a fuzzy lattice modular homomorphicism.
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-15. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.