Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks268 from public data
- 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.
- Graph Neural Networks for Graphs With Heterophily: A Survey343
A systematic taxonomy that governs existing heterophilic GNN models for heterophilic graphs is proposed, aiming to facilitate the development of more effective GNNs across a spectrum of practical applications and learning tasks in the graph research community.
- Heterogeneous Graph Neural Network via Attribute Completion228
It is held that missing attributes can be acquired by a learnable manner, and a general framework for Heterogeneous Graph Neural Network via Attribute Completion (HGNN-AC), including pre-learning of topological embedding and attribute completion with attention mechanism is proposed.
- Semantic Community Identification in Large Attribute Networks211
A novel nonnegative matrix factorization (NMF) model with two sets of parameters, the community membership matrix and community attribute matrix is proposed and the use of node attributes improves upon community detection and provides a semantic interpretation to the resultant network communities.
- 197
- Graph Convolutional Networks Meet Markov Random Fields: Semi-Supervised Community Detection in Attribute Networks149
This work proposes to integrate the techniques of GCN and MRF to solve the problem of semi-supervised community detection in attributed networks with semantic information and exploits both network topology and node semantic information in a complete end-to-end deep network architecture.
- Powerful Graph Convolutional Networks with Adaptive Propagation Mechanism for Homophily and Heterophily140
A novel propagation mechanism is designed, which can automatically change the propagation and aggregation process according to homophily or heterophily between node pairs, and it is theoretically proved that the model can constrain the similarity of representations between nodes according to theirhomophily degree.
- Multi-Intention Oriented Contrastive Learning for Sequential Recommendation128
This work formalizes the denoising problem by selecting the user's main intention, and proposes a novel framework, namely Multi-Intention Oriented Contrastive Learning Recommender (IOCRec), which fuse local and global intentions to unify sequential patterns and intent-level self-supervision signals.
- Topology Optimization based Graph Convolutional Network114
A novel Topology Optimization based Graph Convolutional Networks (TO-GCN) to fully utilize the potential information by jointly refining the network topology and learning the parameters of the FCN is proposed.
- 104
- On Proximity and Structural Role-based Embeddings in Networks102
The misconceptions and key differences between structural roles and communities are clarified, and the general mechanisms (e.g., random walks and feature diffusion) that give rise to community- or role-based structural embeddings are formalized.
- 97
- 94
- 94
- 85
- Semi-supervised community detection based on non-negative matrix factorization with node popularity83
The proposed SSNMF model is improved by introducing the node popularity, namely PSSNMF, which helps to utilize the prior information more effectively and improves the accuracy of community detection, especially on networks with large degree heterogeneity and unbalanced community structure.
- Identifying overlapping communities as well as hubs and outliers via nonnegative matrix factorization76
A novel model where vertices are measured by their centrality in communities and defined as an optimization problem, calculated by nonnegative matrix factorization is proposed, which demonstrates its ability of identifying overlapping communities, hubs and outliers and validate its superior performance in terms of clustering quality.
- Trafformer: Unify Time and Space in Traffic Prediction75
A new model named Trafformer is proposed, which unifies spatial and temporal information in one transformer-style model and outperforms existing methods and provides a promising future direction for the spatial-temporal traffic prediction problem.
- Incorporating network structure with node contents for community detection on large networks using deep learning74
A new approach to seamlessly combine the models of modularity and normalized-cut via the autoencoder is proposed and can provide a nonlinearly deep representation for a large-scale network and reached an efficient community detection.
- Complex Network Clustering Algorithms65
This paper reviews the background, the motivation, the state of arts as well as the main issues of existing works related to discovering network communities, and tries to draw a comprehensive and clear outline for this new and active research area.
- 64
- 60
- Dual-site fluorescent probe for multi-response detection of ClO− and H2O2 and bio-imaging60
The first dual-site fluorescent probe (Geisha-1) for the quantitative detection of ClO- and H2O2 is reported, constructed by chemically grafting N,N-dimethylthiocarbamate and borate to a fluorescence resonance energy transfer (FRET) platform.
- A Unified Weakly Supervised Framework for Community Detection and Semantic Matching55
This paper proposes a unified nonnegative matrix factorization framework simultaneously for community detection and semantic matching by integrating both semi-supervised information and node content and shows that the framework is efficient to match each community with specific semantic information, and the performance are superior over the compared methods.
- 54
- Combined node and link partitions method for finding overlapping communities in complex networks52
A new unified model and method for finding the best overlapping communities on the basis of the associated node and link partitions derived from the same framework is proposed and evaluated on both synthetic and real-world networks with ground-truths.
- Hierarchical community detection with applications to real-world network analysis51
This paper introduces a novel community detection algorithm named PMC, which stands for probabilistically mining communities, to meet the challenging objective of seeking a good trade-off between effectiveness and efficiency in large-scale networks.
- 50
- Dual Intent Enhanced Graph Neural Network for Session-based New Item Recommendation50
A dual-intent network to learn user intent from an attention mechanism and the distribution of historical data respectively is designed, which can simulate users’ decision-making process in interacting with a new item.
- TrustGNN: Graph Neural Network-Based Trust Evaluation via Learnable Propagative and Composable Nature47
A new GNN-based trust evaluation method named TrustGNN, which integrates smartly the propagative and composable nature of trust graphs into a GNN framework for better trust evaluation and significantly outperforms the state-of-the-art methods.
- Identification of hybrid node and link communities in complex networks46
This work introduces a new scheme and an effective approach for identifying complex mixture structures of node and link communities, called hybrid node-link communities, which outperformed the existing methods for finding node or link communities separately.
- Augmenting Affective Dependency Graph via Iterative Incongruity Graph Learning for Sarcasm Detection42
This work explores a novel Iterative Augmenting Affective Graph and Dependency Graph (IAAD) framework to jointly and iteratively learn the incongruity graph structure and demonstrates that the proposed model outperforms state-of-the-art baselines for sarcasm detection with significant margins.
- 42
- Incorporating Network Embedding into Markov Random Field for Better Community Detection42
This work proposes a general Markov Random Field (MRF) framework to incorporate coupling in network embedding which allows better detecting network communities, and improves the accuracy of existing embedding methods, and corrects most wrongly-divided statistically-significant nodes.
- Network-Specific Variational Auto-Encoder for Embedding in Attribute Networks42
A network-specific VAE for network embedding is developed based on a Gaussian mixture model and the block model to reconstruct network structures and demonstrates superior performance over the state-of-the-art methods.
- Modeling with Node Degree Preservation Can Accurately Find Communities42
This work adopted a random-graph null model to faithfully capture community structures by preserving in the model the expected node degrees, and learned using nonnegative matrix factorization is more accurate and robust in representing community structures than the existing methods.
- 41
- 41
- Community Detection via Joint Graph Convolutional Network Embedding in Attribute Network40
This work proposes an unsupervised model for community detection via joint GCN embedding, i.e. JGE-CD, and employs GCN as the basic structure of encoder to match the above two sources of information, and introduces a community detection module considering the community properties into the joint learning process.
- 39
- From Community to Role-based Graph Embeddings.38
This manuscript seeks to clarify the differences between roles and communities, and formalize the general mechanisms that give rise to community or role-based embeddings, and shows mathematically why embedding methods based on these identified mechanisms are either community orrole-based.
- Toward Understanding and Evaluating Structural Node Embeddings37
It is found that degree distributions within nodes’ local neighborhoods can lead to simple yet effective baselines in their own right and guide the future development of structural embedding.
- Autonomous overlapping community detection in temporal networks: A dynamic Bayesian nonnegative matrix factorization approach37
This paper proposes a Bayesian probabilistic model, named Dynamic Bayesian Nonnegative Matrix Factorization (DBNMF), for automatic detection of overlapping communities in temporal networks and demonstrates that the DBNMF model has superior performance compared with two widely used methods, especially when the number of communities is unknown and when the network is highly sparse.
- 35
- A Stochastic Model for Detecting Heterogeneous Link Communities in Complex Networks35
A stochastic model is proposed, which not only describes the structure of link communities, but also considers the heterogeneous distribution of community sizes, a property which is often ignored by other models.
- 34
- ModMRF: A modularity-based Markov Random Field method for community detection33
A novel Markov Random Field (MRF) method is proposed by formalizing modularity as an energy function based on the rich structures of MRF to represent properties and constraints of this problem, and use the max-sum BP to infer model parameters.
- 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.
- Link Community Detection Using Generative Model and Nonnegative Matrix Factorization32
This work proposes a generative model, which is based on the importance of each node when forming links in each community, to describe the structure of link communities, and extends the above method by introducing a strategy of iterative bipartition.
- 29
- A Generalized Deep Markov Random Fields Framework for Fake News Detection28
A graph-theoretic framework that inherits the capability of deep learning while at the same time exploiting the correlations among the news articles (including labeled and unlabeled data), and develops an Ambiguity Learning Guided MRF model as a concretization of GDMRFF.
- 28
- Glucosamine modified near-infrared cyanine as a sensitive colorimetric fluorescent chemosensor for aspartic and glutamic acid and its applications28
A glucosamine modified near-infrared cyanine fluorescent probe has been synthesized and developed as an efficient and visual sensor for the trace level determination of glutamic acid (Glu) and aspartic acid (Asp) with no interference by other amino acids.
- Genetic Algorithm with a Local Search Strategy for Discovering Communities in Complex Networks28
Experimental results show that GALS is highly effective and efficient for discovering community structure and compared with some presently competing algorithms.
- 27
- Active link selection for efficient semi-supervised community detection27
By connecting uncertain nodes to their community hubs and disconnecting the inter-community edges, one can sharpen the block structure of adjacency matrix more efficiently than randomly labeling links as the existing methods did.
- 26
- LLGformer: Learnable Long-range Graph Transformer for Traffic Flow Prediction26
A new model, LLGformer, is introduced, which improves upon traditional Transformer-style models, facilitating more efficient learning of traffic flow data by integrating long-range historical information and two optimization strategies to further boost the speed of training and inference.
- Perseus26
Perseus is presented, a large-scale system that enables the comprehensive analysis of large graphs by supporting the coupled summarization of graph properties and structures, guiding attention to outliers, and allowing the user to interactively explore normal and anomalous node behaviors.
- Contrastive Graph Similarity Networks25
A comprehensive series of experiments indicate that CGSim outperforms state-of-the-art baselines on six datasets and significantly reduces the computational cost, which demonstrates the CGSim model’s superiority over other baselines.
- Commonsense Knowledge Enhanced Sentiment Dependency Graph for Sarcasm Detection25
A Commonsense Sentiment Dependency Graph Convolutional Network (CSDGCN) framework is proposed to explicitly depict the role of external commonsense and inconsistent expressions over the context for sarcasm detection by interactively modeling the sentiment and dependency information.
- 25
- Toward Unsupervised Graph Neural Network: Interactive Clustering and Embedding via Optimal Transport25
- 24
- Latent Network Summarization24
Multi-LENS is an inductive multi-level latent network summarization approach that leverages a set of relational operators and relational functions to capture the structure of egonets and higher-order subgraphs, respectively and shows the effectiveness in detecting anomalies and events in the Enron email communication graph and Twitter co-mention graph.
- Detect Overlapping Communities via Ranking Node Popularities24
The novel approach, based upon the idea of ranking node popularities within communities and using a Bayesian method to shrink communities to optimize an objective function based on the stochastic generative model, is evaluated, showing its superior performance over five state-of-the-art methods.
- Adversarial Mutual Information Learning for Network Embedding23
This work uses the mutual information between node attributes and embedding as a reasonable alternative of this encoding mechanism (which is much easier to track) and introduces another mapping mechanism (which is based on GAN) as a competitor into the adversarial learning system.
- Finding Communities with Hierarchical Semantics by Distinguishing General and Specialized topics23
A novel probabilistic generative model that can find community structures more accurately, but also provide two-level semantic interpretation for each community by distinguishing the general and specialized topics of words is proposed.
- Autoencoder Based Community Detection with Adaptive Integration of Network Topology and Node Contents23
An adaptive method for community detection is presented which is reached by a graph regularized autoencoder approach and introduces a novel adaptive parameter to achieve robust integration of the topological and content information when there exists the mismatch between those two types of information in term of communities.
- Local-Global Defense against Unsupervised Adversarial Attacks on Graphs22
This work proposes an unsupervised defense method that combines local and global defense to improve the robustness of representation and presents a method of attention-based protection against high-risk attacks that penalizes attention coefficients of perturbed edges to encoders.
- BiTe-GCN: A New GCN Architecture via Bidirectional Convolution of Topology and Features on Text-Rich Networks22
BiTe-GCN is proposed, a novel GCN architecture modeling via bidirectional convolution of topology and features on text-rich networks that outperforms the state-of-the-arts by a breakout improvement and can be applied to several e-commerce search scenes such as JD searching.
- Community Detection in Social Networks Considering Topic Correlations22
Network contents including node contents and edge contents can be utilized for community detection in social networks and the topic of each community can be extracted as its semantic information.
- Combination of links and node contents for community discovery using a graph regularization approach22
A novel community detection method under the framework of nonnegative matrix factorization (NMF) is introduced, and the idea that two nodes with similar content will be most likely to belong to the same community to achieve the incorporation of links and node contents is adopted.
- Integrative Network Embedding via Deep Joint Reconstruction22
This paper develops a weight-free multi-component network embedding approach by network reconstruction via a deep Autoencoder that demonstrates superior performance over nine state-of-the-art methods for embedding.
- Similarity-based Regularized Latent Feature Model for Link Prediction in Bipartite Networks22
The proposed framework for link prediction in bipartite networks is called Similarity Regularized Nonnegative Matrix Factorization (SRNMF), which explicitly takes the local characteristics into consideration and encodes the geometrical information of the networks by constructing a similarity based matrix.
- A model framework for the enhancement of community detection in complex networks22
This work proposes a model framework which integrates the enhancement into the whole community detection procedure, and presents a stochastic model to describe the community memberships of nodes and a nonnegative matrix factorization approach to learn the model parameters.
- Does GCL Need a Large Number of Negative Samples? Enhancing Graph Contrastive Learning with Effective and Efficient Negative Sampling21
A new method called GCL with Effective and Efficient Negative samples, E2Neg, is proposed, which learns discriminative representations using only a very small set of representative negative samples, which significantly reduces computational overhead and speeds up model training.
- FUG: Feature-Universal Graph Contrastive Pre-training for Graphs with Diverse Node Features21
An encoder with contrastive constraints to emulate PCA’s generation of basis transformation matrix, which is utilized to losslessly adapt features in different datasets, and introduced a global uniformity constraint to replace negative sampling.
- 21
- Smart Roles21
This paper proposes EMBER, or EMBedding Email-based Roles, which finds email-centric embeddings of network nodes to be used in professional role inference tasks, and automatically captures behavioral similarity between employees in the email network, leading toembeddings that naturally distinguish employees of different hierarchical roles.
- A near-infrared fluorescent probe for rapid and selective detection of hydrosulfide and imaging in live cells21
A heptamethine cyanine-based near-infrared (NIR) fluorescent probe Cy-1 for hydrosulfide detection was developed, with the response mechanism based on nucleophilic substitution of acryloyl moiety by HS− demonstrating superiority in selectivity over Cys.
- Heterogeneous Graph Neural Networks using Self-supervised Reciprocally Contrastive Learning20
A robust heterogeneous graph contrastive learning approach, namely HGCL, is developed, which introduces two views on respective guidances of node attributes and graph topologies and integrates and enhances them by a reciprocally contrastive mechanism to better model heterogeneous graphs.
- 20
- Semi-supervised community detection on attributed networks using non-negative matrix tri-factorization with node popularity20
A semi-supervised non-negative matrix tri-factorization model with node popularity (i.e., PSSNMTF) is designed to detect communities on the topology of the network and node contents are integrated into the PSS NMTF model to find the semantic communities more accurately, namely PSSNMTFC.
- Fast algorithms for core maximization on large graphs19
This paper revisits the challenging yet important problem of core maximization, and proposes a novel algorithm FastCM+ based on several fast search strategies that finds a good-quality answer for edge insertions.
- Deep transfer learning for multi-source entity linkage via domain adaptation19
AdaMEL is a deep transfer learning framework that learns generic high-level knowledge to perform multi-source entity linkage, and leverages the massive unlabeled data from new data sources through domain adaptation to make it generic and data-source agnostic.
- 19
- 19
- ANT COLONY OPTIMIZATION WITH A NEW RANDOM WALK MODEL FOR COMMUNITY DETECTION IN COMPLEX NETWORKS19
Experimental result has shown that MACO is highly effective for discovering communities, and has been evaluated both on synthetic benchmarks and on some real-world networks, and compared with some present competing algorithms.
- 19
- Functional brain abnormalities in major depressive disorder using a multiscale community detection approach18
A novel multiscale community detection method to compare the differences in brain regions between normal controls (NC) and MDD patients is proposed and can provide valuable insights into the diagnosis and treatment of MDD.
- Graph Neural Network for Higher-Order Dependency Networks18
A new end-to-end GNN framework for dealing with higher-order networks directly in the model and the higher- order dependency is used as the neighbor aggregation controller when the node is embedded and updated.
- A Sensitive Colorimetric and Ratiometric Chemosensor for Trivalent Metal Cations18
A novel hydroxyethyl piperazine functionalized cyanine derivative presents selective colorimetric as well as ratiometric absorption responses to trivalent metal cations over a variety of divalent and monovalentMetal cations in 3:7 ethanol-water solution.
- Ant Colony Optimization Based on Random Walk for Community Detection in Complex Networks18
This paper addresses the network clustering problem with a novel ant colony optimization strategy, named as RWACO, which improves prior results on the NCPs and does not require knowledge of the number of communities present on a given network.
- 18
- 17
- BERT2DNN: BERT Distillation with Massive Unlabeled Data for Online E-Commerce Search17
By making the data processing and model training source code public, it is hoped the techniques presented here can help reduce energy consumption of the state of the art Transformer models and also level the playing field for small organizations lacking access to cutting edge machine learning hardwares.
- 17
- 17
- Feature-Structure Adaptive Completion Graph Neural Network for Cold-start Recommendation16
This work proposes a new method called Feature-Structure Adaptive Completion Graph Neural Network (FS-GNN), which is designed to tackle the cold-start problem by simultaneously addressing the missing feature and structure information in a bipartite graph composed of users and items.
- 16
- The (un)supervised NMF methods for discovering overlapping communities as well as hubs and outliers in networks16
This work defines a novel generative model and formalizes the detection of overlapping communities as well as hubs and outliers as an optimization problem on it and presents a community detection method, which detects all three structures altogether without prior knowledge of the number of communities.
- Topic Enhanced Sentiment Spreading Model in Social Networks Considering User Interest15
A unified probabilistic framework is developed to formalize the problem into a topic-enhanced sentiment spreading model that significantly outperforms several alternative methods in predicting users' sentimental status and discovers an intriguing phenomenon that positive and negative sentiment is more relevant to user interest than neutral ones.
- 15
- Improving the Efficiency and Effectiveness of Community Detection via Prior-Induced Equivalent Super-Network15
This work proposes a general, high-speed, effective and parameter-free semi-supervised community detection framework that transforms the original network into an equivalent but much smaller Super-Network, which perfectly ensures the must-link constraints and effectively encodes cannot- link constraints.
- 15
- A Stochastic Model for Detecting Overlapping and Hierarchical Community Structure15
This work proposes a generative model that employs a nonnegative matrix factorization formulization with a l2,1 norm regularization term, balanced by a resolution parameter, and derives the multiplicative update rule to learn the model parameters, and offers the proof of its correctness.
- 14
- GCN for HIN via Implicit Utilization of Attention and Meta-Paths14
A novel neural network method via implicitly utilizing attention and meta-paths is proposed, which can relieve the severe overfitting brought by the current over-parameterized attention mechanisms on HIN.
- Budget-constrained Truss Maximization over Large Graphs14
A component-based dynamic programming algorithm is developed for enlarging k-truss mostly, which makes a balance of budget assignment and inserts multiple edges simultaneously into all (k-1)-light components.
- A Time-Sensitive Graph Neural Network for Session-Based New Item Recommendation13
TSGNN provides a novel time-sensitive session graph constructing technique to solve the first problem and designs a time-aware attention mechanism to accurately discern user preferences, which recommends items with high relevance scores for users.
- 13
- Modeling with Node Popularities for Autonomous Overlapping Community Detection13
A stochastic model is presented to accommodate the relative importance and the expected degree of every node in each community and every overlapping community is inferred by ranking the nodes according to their importance.
- Decentralized mining social network communities with agents13
A multi-agent based decentralized algorithm, in which a group of autonomous agents work together to mine a network through a proposed self-aggregation and self-organization mechanism, potentially suitable for dealing with distributed networks, whose global structures are hard to obtain due to their geographical distributions, decentralized controls or huge sizes.
- Ant Colony Optimization with Markov Random Walk for Community Detection in Graphs13
A new ant colony optimization strategy, named as ACOMRW, is presented, which improves prior results on the NCP problem and does not require knowledge of the number of communities present on a given network.
- TGFormer: Towards temporal graph transformer with auto-correlation mechanism12
This work proposes TGFormer, a novel Transformer architecture specifically designed for temporal graphs that redefines temporal graph learning by establishing a trajectory framework that aligns with time series analysis principles and develops an auto-correlation mechanism that systematically uncovers periodic dependencies in node interactions.
- Towards Global-Topology Relation Graph for Inductive Knowledge Graph Completion12
TARGI, a simple yet effective inductive method for KGC that accurately learns the topological structures and embeddings of new relations, and guides the embedding learning of new entities, outperforms 15 SOTA methods.
- Multi-Modal Sarcasm Detection Based on Dual Generative Processes12
This work proposes a multimodal sarcasm detection method based on dual generative processes that explicitly model the modalities' contributions based on prediction performance and autonomously adjust the weight distribution.
- 12
- Contribution of Chinese herbal medicine in the treatment of coronavirus disease 2019 ( COVID ‐19): A systematic review and meta‐analysis of randomized controlled trials12
It is shown that the CHM group improved in lung CT, clinical cure rate, clinical symptom score and time to negative for viral nucleic acid, and therefore, CHM may be used as a potential candidate for COVID‐19.
- Community Detection in Social Networks Considering Social Behaviors12
A generative community detection model SBCD (social behavior-based community detection) is proposed by combining network topology and content, in which the above social behaviors play a core role.
- 12
- Robust Detection of Communities with Multi-semantics in Large Attributed Networks12
The transition probability which is an important concept in Markov chain is introduced into a well-designed nonnegative matrix factorization framework and this new transition probability matrix with a suitable prior which plays the role of depicting the relationship between communities and topics can perform well in this task.
- Constrained common cluster based model for community detection in temporal and multiplex networks12
A constrained common cluster based model ( C 3 model) is proposed to analyze the temporal and multiplex networks, which can not only detect the community structure, but also identify the importance of each node based on the common cluster structure of both two classes of networks.
- Extending a configuration model to find communities in complex networks12
This work generalizes a configuration model, namely, a null model of modularity, to solve the problem of discovery of communities in complex networks, and combines both the community property and degree sequence preserving into a single unified model, which gives better community results compared with other models.
- 12
- Dynamic Neighborhood Modeling via Node-Subgraph Contrastive Learning for Graph-Based Fraud Detection11
This work presents a novel dynamic neighborhood modeling via node-subgraph contrastive learning for graph-based fraud detection, dubbed DCL-GFD, and designs a node abnormality estimation module from the perspective of feature, which analyses the likelihood of a node belonging to fraud or benign by comparing the feature similarity between the target node and its corresponding subgraph.
- Environmental enrichment enhanced neurogenesis and behavioral recovery after stroke in aged rats11
Environmental enrichment improved post-stroke recovery on the basis of enhancing neurogenesis in aged rats and suppressed apoptosis and improved the motor functions after stroke in the aged rats.
- 11
- AN ANT-BASED ALGORITHM WITH LOCAL OPTIMIZATION FOR COMMUNITY DETECTION IN LARGE-SCALE NETWORKS11
Experimental results demonstrate that MABA is both effective and efficient (in near linear time with respect to the size of network) for discovering communities.
- Fast convergent key distribution algorithms using a dual quantum channel11
Three new algorithms over a dual quantum channel to enhance the convergence speed of the key distribution process and improve the efficiency of thekey distribution process as well as deter Eve are presented.
- Integrating Co-Training with Edge Discrimination to Enhance Graph Neural Networks Under Heterophily10
It is argued that identifying neighbors with different class labels and exploiting them individually is crucial for heterophilic GNNs, and a simple and efficient novel co-training approach, EG-GCN, which uses group aggregation to handle homophilic and heterophilic neighbors separately is proposed.
- 10
- 10
- 10
- Autonomous Semantic Community Detection via Adaptively Weighted Low-rank Approximation10
This work develops a novel NMF-based method to autonomously determine the number of semantic communities and community structure simultaneously, showing its superior performance over several competing methods on synthetic and large real-world social networks.
- Fast Complex Network Clustering Algorithm Using Agents10
A fast network clustering algorithm in which each node is regarded as an agent, and each agent tries to maximize its local function in order to optimize network modularity defined by function Q, rather than optimize function Q from the global view as traditional methods.
- A Dynamic Knowledge Update-Driven Model with Large Language Models for Fake News Detection9
A dynamic knowledge update-driven model for fake news detection (DYNAMO) is proposed, which leverages knowledge graphs to achieve continuous updating of new knowledge and integrates with large language models to fulfill dual functions: news authenticity detection and verification of new knowledge correctness, solving the two key problems of ensuring the authenticity of new knowledge and deeply mining news semantics.
- 9
- 9
- 9
- Attentional factorization machine with review-based user–item interaction for recommendation9
An attentional factorization machine with review-based user–item interaction for recommendation (AFMRUI) is proposed, which first leverages RoBERTa to obtain the embedding feature of each user/item review, and combines bidirectional gated recurrent units with attention network to highlight more useful information in both user and item reviews.
- Emotional Contagion-Based Social Sentiment Mining in Social Networks by Introducing Network Communities9
A new generative model called Community-Enhanced Social Sentiment Mining (CESSM), which considers the emotional contagion between writers and readers to capture precise social sentiment, and incorporates network communities to capture coherent topics.
- An End-to-End Community Detection Model: Integrating LDA into Markov Random Field via Factor Graph9
This paper integrates LDA into MRF to form an end-to-end learning system where their parameters can be trained jointly, and shows that this approach compares favorably with the state-of-the-art methods.
- 9
- HeterGP: Bridging Heterogeneity in Graph Neural Networks with Multi-View Prompting8
This work introduces a novel prompting strategy named HeterGP tailored for highly heterophilic scenarios, and presents a dual-view approach to capture both homophilic and heterophilic information, along with a prompt graph design that encompasses token initialization and insertion patterns.
- A Survey on Temporal Interaction Graph Representation Learning: Progress, Challenges, and Opportunities8
A comprehensive taxonomy of state-of-the-art TIGRL methods is proposed, systematically categorizing them based on the types of information utilized during the learning process to address the unique challenges inherent to TIGs.
- TeKo: Text-Rich Graph Neural Networks With External Knowledge8
This work presents a novel text-rich GNN with external knowledge (TeKo), and devise a reciprocal convolutional mechanism for the constructed heterogeneous semantic network, enabling network structure and textual semantics to collaboratively enhance each other and learn high-level network representations.
- On Generalizing Static Node Embedding to Dynamic Settings8
This paper introduces the notion of an ε-graph time-series that uses a fixed number of edges for each graph, and shows its superiority in practical settings over the standard solution, and finds that the new dynamic embedding methods based on this framework perform comparably and on average better than the state-of-the-art embedding Methods designed specifically for temporal graphs in link prediction tasks.
- A Novel Generative Topic Embedding Model by Introducing Network Communities8
A new generative topic embedding model is given which incorporates documents (with topics) and network (with communities) together, and uses probability transition to describe the relationship between topics and communities to make it robust when Topics and communities do not match.
- 8
- PERSEUS-HUB: Interactive and Collective Exploration of Large-Scale Graphs8
Perseus-Hub is a large-scale graph mining tool that computes a set of graph properties in a distributed manner, performs ensemble, multi-view anomaly detection to highlight regions that are worth investigating, and provides users with uncluttered visualization and easy interaction with complex graph statistics.
- Exploring the roles of cannot-link constraint in community detection via Multi-variance Mixed Gaussian Generative Model8
This is the first work on discovering and exploring the importance of cannot-link constraints in semi-supervised community detection and proposes to characterize this process as a Multi-variance Mixed Gaussian Generative (MMGG) Model to address diverse degrees of confidences that exist in network topology and pairwise constraints and formulate it as a weighted nonnegative matrix factorization problem.
- 8
- 7
- Embedding text-rich graph neural networks with sequence and topical semantic structures7
AS-GNN, an end-to-end adaptive GNN architecture via unified modelling of semantic structure and network propagation on text-rich networks is proposed, which adopts the cascaded “Structure- GNN” pattern to promote the efficiency of the model and be easily combined with existing GNNs.
- TransFusion: Multi-Modal Fusion for Video Tag Inference via Translation-based Knowledge Embedding7
This paper proposes TransFusion, an end-to-end supervised learning framework that fuses multi-modal embeddings with the knowledge embedding to derive the video representation and adopts a dual attentive approach to learn both the modality importance in fusion and relation importance in inference.
- Self-Guided Community Detection on Networks with Missing Edges7
This paper proposes a community self-guided generative model which jointly completes the edges-missing network and identifies communities and developed an effective model inference method that combines a nested Expectation-Maximization algorithm and Metropolis-Hastings Sampling.
- Anti-collision Algorithm based on Multi-threaded RFID Lock Position7
From the test results, compared with the original algorithm, this algorithm reduces the times of collision identification and digit number of data transmission, and improves the system performance.
- 6
- 6
- 6
- Better value estimation in Q-learning-based multi-agent reinforcement learning6
This paper introduces the double critic networks structure and the delayed policy update to Q-learning-based multi-agent MARL methods, which reduce the overestimation and enhance the quality of policy updating.
- 6
- 6
- 6
- 6
- Morphological Verb-Aware Tibetan Language Model6
This paper proposes a morphological verb-aware TLM by offline learning via a character frequency reweighting strategy and online tuning of discriminative weights conditioned on morphological verbs.
- 6
- 6
- Joint Domain Adaptive Graph Convolutional Network5
A joint adversarial domain adaptive graph convolutional network (JDA-GCN) is proposed that is uniquely augmented with structural graph alignment, so as to enhance the efficacy of knowledge transfer and mark a significant advancement over existing state-of-the-art graph domain adaptation algorithms.
- 5
- 5
- TMS: Temporal multi-scale in time-delay neural network for speaker verification5
Experimental results show that the TMS-method-based model outperformed state-of-the-art ASV models and improved robustness and achieved a 29%–46% increase in the inference speed compared to ECAPA-TDNN.
- Inflation Improves Graph Neural Networks5
This paper first finds that the feature propagation process of GNNs can be taken as a Markov chain, and then introduces a new graph inflation layer into GNN's to prevent noise propagating from local neighbourhoods to the whole graph with the increase of network layers.
- Improve emotional speech synthesis quality by learning explicit and implicit representations with semi-supervised training5
Experiments show that the proposed novel model to improve emotional speech synthesis quality by learning explicit and implicit representations with semi-supervised learning can improve the naturalness and expressiveness of synthetic speech and is better than the baseline model.
- Fast Flow-based Random Walk with Restart in a Multi-query Setting5
This work introduces a new, intuitive two-step divide-and-conquer formulation and a corresponding parallelizable method, FlowR, for solving RandomWalk with Restart with two goals: fast and accurate computation under multiple queries and one-time message exchange between subproblems.
- 5
- 5
- The (un)supervised detection of overlapping communities as well as hubs and outliers via (bayesian) NMF5
A novel model to identify overlapping communities as well as hubs and outliers is defined and a normalized symmetric nonnegative matrix factorization algorithm is proposed to learn the parameters of the model.
- Universal Graph Self-Contrastive Learning4
A novel GCL framework called GRAph learning via Self-contraSt (GRASS) is introduced, which implicitly ensures accurate node-node contrast by capturing high-hop co-inclusion relationships, thereby enabling GRASS to be universally applicable to graphs with varying degrees of homophily.
- Adaptive Multi-Channel Deep Graph Neural Networks4
An adaptive multi-channel deep graph neural network (AMD-GNN) to adaptively and symmetrically aggregate information from the deep receptive field to achieve state-of-the-art performance on node classification tasks with deep models.
- SOT: Self-supervised Learning-Assisted Optimal Transport for Unsupervised Adaptive Speech Emotion Recognition4
A self-supervised learning (SSL)- assisted optimal transport (SOT) algorithm for cross-domain SER is proposed and experimental results showed that SOT dramatically outperformed state-of-the-art UDAs.
- TMS: A Temporal Multi-scale Backbone Design for Speaker Embedding4
The performance of the new TMS method for automatic speaker verification (ASV) on in-domain and out-of-domain conditions is investigated and results show that the TMS-based model obtained a significant increase in the performance over the SOTA ASV models, which had a faster inference speed.
- IvySyn: Automated Vulnerability Discovery in Deep Learning Frameworks4
IvySyn leverages the statically-typed nature of native APIs in order to automatically perform type-aware mutation-based fuzzing on low-level kernel APIs in order to automatically perform type-aware mutation-based fuzzing on low-level kernel APIs.
- Guest Editorial: Graph-powered machine learning in future-generation computing systems4
A set of research topics such as knowledge graph reasoning, graph self-supervised learning, temporal graph modeling, and graph embedding techniques have emerged, and applications such as graph-based anomaly detection, social recommendation, social influence analytics are becoming important issues for the research community.
- 4
- 4
- 4
- Towards Graph Foundation Model: Node Feature Transfer Invariant Modeling on General Graphs3
A Transfer-Invariant Graph (TIG) foundational model is introduced to transform features of different dimensions into a unified structural representation that can effectively learn and extract TIM and intrinsic graph knowledge by self-supervised learning.
- 3
- Predictive models of epidermal growth factor receptor mutation in lung adenocarcinoma using PET/CT‐based radiomics features3
The objective of this investigation is to construct and validate predictive models for EGFR mutations in LAC by leveraging PET/CT‐derived radiomics features, thereby refining diagnostic precision and facilitating tailored treatment strategies.
- 3
- 3
- 3
- Community Detection on Social Networks With Sentimental Interaction3
A new Community Sentiment Diffusion Detection Model (CSDD), which utilizes sentimental information embedded in forward posts is proposed and an efficient variational algorithm for model inference is presented.
- Learning Neighbor User Intention on User–Item Interaction Graphs for Better Sequential Recommendation3
A Neighbor user Intention-based Sequential Recommender (NISRec), which utilizes the intentions of high-order connected neighbor users as high-order collaborative signals in order to improve recommendation performance for the target user.
- 3
- Leveraging the Graph Structure of Neural Network Training Dynamics3
This work proposes a compact, expressive temporal graph framework that effectively captures the dynamics of many workhorse architectures in computer vision and demonstrates that the proposed framework captures generalizable dynamics as summaries extracted from smaller-width networks are effective when evaluated on larger widths.
- A Network Embedding-Enhanced NMF Method for Finding Communities in Attributed Networks3
This study identifies a novel non-negative matrix factorization method which both employs network embedding to enhance the representation power of network topology for communities and also integrates networkTopology and node contents for further raising the quality of community detection.
- Exponent and Logarithm Component-Wise Construction Method of FCM Clustering Validity Function Based on Subjective and Objective Weighting3
The feasibility and effectiveness of the proposed clustering validity function construction method are proved and the simplest structure and the best classification effect was selected by comparison.
- 3
- 3
- 3
- TeleLink: Link Prediction in Social Network Based on Multiplex Cohesive Structures3
This work presents TeleLink, a general link prediction model that works for networks with single and multiple relationships, and extends TeleLink to a semi-supervised scheme, incorporating partially labeled information.
- Erratum: Identifying overlapping communities as well as hubs and outliers via nonnegative matrix factorization3
A novel model where vertices are measured by their centrality in communities is proposed, and the identification of overlapping communities, hubs, and outliers is defined as an optimization problem, calculated by nonnegative matrix factorization.
- FUZZY NEURAL NETWORK'S APPLICATION IN FURNACE TEMPERATURE COMPENSATION BASED ON ROLLING INFORMATION FEEDBACK3
FNN is used to deal with the feedback of rolling information, and then real-time compensation of furnace temperature setting can be obtained, and energy consumption of hot rolling process can be reduced greatly, and rolling security of roughing mill can be guaranteed at the same time.
- 2
- Integrated Mixture of Neighborhood and Community Experts for Graph-Based Fraud Detection2
This work proposes NeCo, a novel approach that integrates mixture of neighborhood and community experts for graph-based fraud detection, and introduces a fraud-discriminative representation preservation mechanism from a neighborhood perspective, leveraging the empirical finding that fraud nodes tend to exhibit larger feature propagation discrepancies compared to benign nodes.
- 2
- 2
- 2
- 2
- CoBjeason: Reasoning Covered Object in Image by Multi-Agent Collaboration Based on Informed Knowledge Graph2
The proposed CoBjeason model could offer novel insights into more basic Computer Vision tasks, such as Semantic Segmentation with better understanding on the current scene when some objects are blurred or covered, Visual Question Answering with enhancement on the inference in more complicated visual context when some objects are covered or invisible, and Image Caption Generation with the augmentation on the richness of visual context for images containing partially visible objects.
- InjectTST: A Transformer Method of Injecting Global Information into Independent Channels for Long Time Series Forecasting2
An injection method for global information into channel-independent Transformer, InjectTST, which retains the channel-independent backbone and gradually inject global information into individual channels in a selective way and can achieve stable improvement compared with state-of-the-art models.
- 2
- 2
- 2
- 2
- Layer Clustering-Enhanced Stochastic Block Model for Community Detection in Multiplex Networks2
Results on two real-world datasets suggest that the proposed SBM-based MN-community-detection algorithm can mine meaningful relationships between network layers and can be more accurate and robust than multilayer-Louvain algorithm.
- 3-in-1 Correlated Embedding via Adaptive Exploration of the Structure and Semantic Subspaces2
A novel generative model is proposed to formulate the generation process of the network and content from the embeddings, with respect to the Bayesian framework, and automatically leans to the information which is more discriminative.
- 2
- Unified Multi-Domain Graph Pre-training for Homogeneous and Heterogeneous Graphs via Domain-Specific Expert Encoding1
It is empirically demonstrated that a balanced mixture of homogeneous and heterogeneous graph pre-training benefits downstream tasks and a unified multi-domain Graph Pre-training method across Homogeneous and Heterogeneous graphs (GPH2) is proposed.
- Unveiling Backdoor Propagation in Graphs: Neuron-Centric Defense Mechanisms1
This work designs a novel reverse engineering technique to identify triggers that activate backdoor neurons, and eliminate their harmful effects by asymmetric unlearning and recovering at the neuron level, and proposes a novel defense called Graph Backdoor Neuron Defense (GBND).
- 1
- 1
- 1
- 1
- 1
- 1
- A Simple and Effective Community Detection Method Combining Network Topology with Node Attributes1
This work proposes a simple while effective approach which utilizes attribute information to partition nodes into communities by maximizing network modularity and demonstrates the superior performance of the new approach over some state-of-the-art approaches.
- 1
- 1
- Security in Computing and Communications: Second International Symposium, SSCC 2014, Delhi, India, September 24-27, 2014. Proceedings1
This paper shows that the proposed biometricbased user authentication scheme for wireless sensor networks has some security pitfalls such as it is not resilient against node capture attack, it is insecure against impersonation attack, and it is secure against man-in-the-middle attack.
- 1
- Research on Cluster Aggregation Approaches1
This paper made an overview of the relevant knowledge of the clustering aggregation, presented the concepts and advantages, and analysed three approaches to get clustering members according to the uesd cluster algorithms.
- Selective approach for neural network ensemble based on network clustering technology and its application1
Test results reveal that the proposed neural network ensemble approach is better than the traditional fertilization models, linear weighted ensemble model and nonlinear ensemble model, and the ability of generalization of the approach is strong.
- IVQ-GNN: Mitigating Performance Gap from Graph Connection Pattern Inconsistency via Vector Quantization–
This paper delves deeper into heterophily and homophily by shifting from coarse-grained heterophily ratios to a unified, fine-grained formulation based on connection patterns, and reveals an uneven distribution and a train–test gap of these patterns.
- MDCD: Meta-Learning Driven Conditional Diffusion for User Cold-Start in Conversational Recommendation–
This work proposes an innovative meta-learning-driven conditional diffusion model that uses a generative model to directly model the data distribution and learn personalized user patterns, enhancing recommendation stability and the meta-learning framework implicitly constrains the optimization process of the diffusion model and other components, enhancing the model's ability to rapidly adapt to new users and dynamic preferences.
- Collateral Damage Constrained Backdoor Attacks on Graph Neural Networks–
The proposed Collateral Damage Constrained Graph Backdoor Attack (CDCA) combines neighborhood-aware target node selection with a self-constrained trigger generation strategy to suppress trigger-induced propagation by enforcing prediction consistency on clean K-hop neighboring nodes.
- –
- –
- –
- –
- When Evidence Falls Short: Router-Guided Fake News Detection with Pattern Augmentation–
A Router-Guided Fake News Detection Framework with Pattern Augmentation (RGPA) is proposed, which guides news to appropriate reasoning paths adaptively based on a multi-dimensional quality assessment, prioritizing high-quality evidence while mitigating noise.
- LLM-FSGNN: LLM-Guided Feature-Structure Augmentation Graph Neural Network for Cold-Start Recommendation–
This work proposes an innovative Large Language Model-Guided Feature-Structure Augmented Graph Neural Network (LLM-FSGNN), a framework that integrates external semantic knowledge with internal graph structures via LLMs and introduces a graph structure enhancement module to complement structural relations from a graph-based perspective.
- A Unified Prompt for Enhancing Heterogeneous Graph Pre-training via Edge-based Message Passing–
This work focuses on the common encoder module shared across pre-trained models and the multi-relational edge structure unique to heterogeneous graphs, proposing a novel heterogeneous graph prompt-tuning method named HGMRP, which significantly outperforms existing heterogeneous prompt methods and validates its effectiveness and superiority.
- –
- –
- JT-Safe: Intrinsically Enhancing the Safety and Trustworthiness of LLMs–
This paper focuses on enhancing pre-training data to improve the trustworthiness and safety of LLMs, and refers to the Data with World Context as DWC, to better anchor pre-training data within real-world scenarios, thereby reducing uncertainty in model training and enhancing the model's safety and trustworthiness.
- Combined with the semantic features of CT and selected clinical variables, a machine learning model for accurately predicting the prognosis of Omicron was established–
This research developed three machine learning models to foresee the results for Omicron-infected patients to provide more possibilities for the identification of the two.
- –
- –
- –
- –
- An Approach for Constructing Service Developers Oriented Heterogeneous Information Network–
Results show that the approach proposed in this paper can identify the implicit relationships among services and developers by making full use of the characteristics of different types of nodes and the interaction information among nodes, consequently provide support for on-demand recommendations of service developers based on similarity measurement.
- Scalable Community Identification with Manifold Learning on Speaker I-Vector Space–
The new model is based on the probabilistic community memberships, and is further refined with the idea that: if two connected nodes have a high similarity, their community membership distributions in the model should be made close, and thus better respects the structure of the underlying manifold than the existing community detection methods.
- –
- Bridging Network Embedding and Graph Summarization.–
Extensive experiments on both synthetic and real-world graphs show that Multi-LENS achieves 2 - 89% improvement in AUC for link prediction, while requiring less than 79x space compared to existing representation learning approaches.
- Design of an Access Control Management System for Buildings Based on CAN–
A comprehensive access control management system which has functions of person-location and speech communication and information notice and so on is proposed which has many features such as flexible network construction and high reliability as well as friendly user interface.
- POSTER: Performance signatures of mobile phone browsers–
A dynamic mobile browser fingerprinting technique that records the browser's behavior and execution characteristics by running background customized browser scripts based on the use of Javascript, HTML5, Flash, and other scripts that are used to generate performance signatures of mobile browsers to detect the browser used, the operating system version, and device type.
- –
- Simulation of data transmission in network environment–
This project tries to simulate data transmission which can be further implemented in secure communication situation by simulating the data transmission in the metro system.
- –
- A New Classification Algorithm Based on Fuzzy C-Means–
An improved algorithm based on fuzzy C-means, a traditional clustering algorithm was finished, which was used to complete the data classification first, and several standard datasets from UCI were tested, which shows a well classification result is found.
- Fast convergent and secure key distribution protocols using a dual-quantum channel–
This dissertation presents three new quantum key distribution protocols over a dual quantum-channel to enhance the convergence speed of the key distribution process and proposes several effective defense mechanisms to frustrate Eve's attack.
- Realization of pure delay system simulation interface based on VB and MATLAB interfacing–
Interfacing programmed of VB and MATLAB was applied to realize man-machine interface function disengaged MATLAB environment and emulation results of PID algorithm,dahlin algorithm and SMITH algorithmic by selection of six-imported signal were obtained.
- –
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.