Query：
学者姓名：郭昆
Refining：
Year
Type
Indexed by
Source
Complex
Former Name
Co-
Language
Clean All
Abstract ：
Many real-world networks can be treated as heterogeneous information networks (HINs) that consist of various types of nodes, like different proteins and molecules in biological networks and different authors and papers in citation networks. Multiple network data mining tasks can be conducted on HINs to capture the complex relationships between multi-type nodes. In recent years, random walk based HIN embedding has drawn increasing attention. Furthermore, the meta-path or meta-graph guided random walk is one of the most widely used techniques in HIN embedding methods. However, existing HIN embedding methods still face several difficulties. Firstly, the meta-paths or meta-graphs often need to be predefined, which relies heavily on domain knowledge and incomplete information coverage. Secondly, these methods treat all relations without distinction, which inevitably limits the capability of HIN embedding. Thirdly, they do not focus on preserving finer-grained meta-graph semantics. In this paper, a HIN embedding algorithm based on adaptive meta-schema considering relation distinction and semantic preservation (HINEAS) is proposed. In order to avoid the selection of meta-paths or meta-graphs, an adaptive meta-schema extraction is designed. In heterogeneous node sequence generation, a biased random walk strategy based on the adaptive meta-schema is presented to embed the different relationships’ influence. Finally, an enhanced embedding strategy based on semantic preservation of the adaptive meta-schema is proposed to effectively extract topology and preserve the meta-graph’s fine-grained semantics. Experiments on real-world datasets show that HINEAS significantly outperforms state-of-the-art methods. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.
Keyword ：
Graph embeddings Graph embeddings Graph theory Graph theory Network embeddings Network embeddings
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Wu, Ling , Gao, Pingping , Lu, Jinlu et al. Heterogeneous Information Network Embedding Based on Adaptive Meta-Schema Considering Relation Distinction and Semantic Preservation [C] . 2025 : 47-63 . |
| MLA | Wu, Ling et al. "Heterogeneous Information Network Embedding Based on Adaptive Meta-Schema Considering Relation Distinction and Semantic Preservation" . (2025) : 47-63 . |
| APA | Wu, Ling , Gao, Pingping , Lu, Jinlu , Guo, Kun , Zhang, Qishan . Heterogeneous Information Network Embedding Based on Adaptive Meta-Schema Considering Relation Distinction and Semantic Preservation . (2025) : 47-63 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
The Multi-Label Propagation Algorithm (MLPA) identifies and reveals community structure by passing labels between network nodes, and is also suitable for dealing with complex networks with overlapping communities. Due to its flexibility and effectiveness, the algorithm has been successfully applied in a number of fields, including image segmentation, text classification, and bioinformatics. In today’s society where personal privacy protection is increasingly important, how to detect communities without revealing sensitive information has become a hot issue in the field of network analysis. Existing privacy-preserving multi-label propagation algorithms primarily rely on anonymization and homomorphic encryption techniques. While homomorphic encryption can protect privacy, the complex encryption and decryption processes incur significant computational costs, making it challenging to achieve efficient computation while ensuring accuracy and privacy. In this paper, we propose a Secure and Efficient Federated Multi-Label Propagation Algorithm (SEFMLPA) that combines an anonymization strategy with a secret sharing strategy, considering the attribute similarities between nodes to ensure privacy, accuracy, and efficiency. The experimental results indicate that SEFMLPA achieves an accuracy comparable to the latest algorithms and reduces runtime by 80%. These significant improvements validate the effectiveness and superiority of our approach. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.
Keyword ：
Anonymity Anonymity Differential privacy Differential privacy Encryption algorithms Encryption algorithms Federated learning Federated learning Privacy-preserving techniques Privacy-preserving techniques
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Guo, Chen , Hu, Wenbin , Xiang, Zhishang et al. Secure and Efficient Federated Multi-label Propagation via Secret Sharing [C] . 2025 : 329-343 . |
| MLA | Guo, Chen et al. "Secure and Efficient Federated Multi-label Propagation via Secret Sharing" . (2025) : 329-343 . |
| APA | Guo, Chen , Hu, Wenbin , Xiang, Zhishang , Dong, Shiyu , Zhang, Qishan , Guo, Kun . Secure and Efficient Federated Multi-label Propagation via Secret Sharing . (2025) : 329-343 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Graphs widely exist in real-world, and Graph Neural Networks (GNNs) have exhibited exceptional efficacy in graph learning in diverse fields. With the strengthening of data privacy protection worldwide in recent years, Federated graph neural networks (FedGNNs) have gained increasing attention in academia and industry owing to their ability to train the model in a collaborative manner while complying with the privacy protection regulations. However, in federated learning, the non-independent and identically distributed (non-IID) problem of local data possessed by multiple participants can significantly undermine model accuracy. We propose a new Decentralized Federated Graph Normalized AutoEncoder (D-FGNAE). First, the model is designed as a decentralized federated learning framework with dynamically assigned tripartite roles. This design eliminates the fixed server role found in traditional federated learning, enhances system fault tolerance, avoids single points of failure, and protects model privacy. Second, the splitting and correcting of calculation by layer in the model, along with the special design of the normalization layer, effectively tackle the non-IID problem in both the structural and attribute aspects. Experimental results on real-world networks demonstrate the effectiveness of D-FGNAE, which can achieve nearly the same accuracy as the centralized model. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.
Keyword ：
Adversarial machine learning Adversarial machine learning Contrastive Learning Contrastive Learning Decentralized control Decentralized control Differential privacy Differential privacy Federated learning Federated learning Graph neural networks Graph neural networks
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Liang, Yuting , Cai, Weixin , Guo, Kun . D-FGNAE: Decentralized Federated Graph Normalized AutoEncoder [C] . 2025 : 281-296 . |
| MLA | Liang, Yuting et al. "D-FGNAE: Decentralized Federated Graph Normalized AutoEncoder" . (2025) : 281-296 . |
| APA | Liang, Yuting , Cai, Weixin , Guo, Kun . D-FGNAE: Decentralized Federated Graph Normalized AutoEncoder . (2025) : 281-296 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Large-scale graphs have become prevalent with the advent of the big data era. Distributed graph computing systems are commonly used for processing and analyzing large-scale graphs, with graph partitioning being a key prerequisite for their efficient computation. Graph partitioning aims to balance the load across partitions while minimizing the number of cut-edges. Moreover, it should achieve high efficiency and scalability. However, the existing popular graph partitioning algorithms do not fully take into account the internal topology of real-world graphs, which affects the final partition quality and convergence. Meanwhile, they easily fall into the local optimum due to partition load constraints. This paper introduces a Novel Optimized Balanced Graph Partitioning algorithm (NOBGP). First, we propose an initialization strategy based on label propagation of core vertices to achieve initial partitions with good locality and accelerate convergence. Second, we optimize the label propagation process to ensure balanced partitions and propose a probability-based disruption strategy to avoid the local optimum. We implement NOBGP on the distributed graph computing framework GraphX. Extensive experimental results on real-world graphs show that the proposed algorithm is scalable and performs better than the existing algorithms. We also run PageRank and Louvain applications using the graph partitioning results to demonstrate the efficiency of our algorithm. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.
Keyword ：
Graph algorithms Graph algorithms Knowledge graph Knowledge graph
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Chen, Jiebin , Hu, Ziqiang , Ye, Renjie et al. NOBGP: A Novel Optimized Balanced Graph Partitioning Algorithm [C] . 2025 : 313-328 . |
| MLA | Chen, Jiebin et al. "NOBGP: A Novel Optimized Balanced Graph Partitioning Algorithm" . (2025) : 313-328 . |
| APA | Chen, Jiebin , Hu, Ziqiang , Ye, Renjie , Zhang, Qishan , Guo, Kun . NOBGP: A Novel Optimized Balanced Graph Partitioning Algorithm . (2025) : 313-328 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Unified stream and batch computing (USBC) aims to incorporate stream and batch computation into a unified framework, thereby enabling the development of a one-stop solution for stream and batch data processing and enhancing the generalization of the framework. However, research on unified graph computing models (UGCMs) faces several challenges. First, existing UGCMs need to consider all graph information in the cache during the incremental update phase, thus leading to decreased execution efficiency. Second, existing UGCMs use fixed bytes to store nodes without considering the actual space occupied by nodes resulting in wasted memory when dealing with large graphs. This paper proposes a UGCM with Local Updates for community detection (UGCM-LU). We first implement a local update strategy to consider partial information of the graph to achieve incremental updates. Secondly, we also designed a byte-compression-based module to store graph data according to the space occupied by nodes. The experimental results show the effectiveness and efficiency of the model in real-world and artificial networks. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.
Keyword ：
Batch data processing Batch data processing Lutetium alloys Lutetium alloys
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Li, Hong , Wu, Ling , Guo, Kun . UGCM-LU: A Unified Stream and Batch Graph Computing Model with Local Update for Community Detection [C] . 2025 : 266-280 . |
| MLA | Li, Hong et al. "UGCM-LU: A Unified Stream and Batch Graph Computing Model with Local Update for Community Detection" . (2025) : 266-280 . |
| APA | Li, Hong , Wu, Ling , Guo, Kun . UGCM-LU: A Unified Stream and Batch Graph Computing Model with Local Update for Community Detection . (2025) : 266-280 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Recently, heterogeneous graph contrastive learning, which can mine supervision signals from the data, has attracted widespread attention. However, most existing methods employ random data augmentation strategies to construct contrastive views, which may destroy the semantic information in heterogeneous graphs. Moreover, they often select positive and negative samples based solely on node-level proximity and overlook hard samples that are difficult to distinguish from anchors. To solve the above problems, we propose a Community-Aware Heterogeneous Graph Contrastive Learning model called CAHGCL. In particular, we design an adaptive data augmentation strategy to construct views, including feature augmentation and topology augmentation. To improve the quality of samples, we propose a dynamic sample weighting strategy based on node similarity and community information, capable of identifying both hard positive samples and hard negative samples. Finally, we introduce community-level contrast to improve community cohesion. Extensive experiments and analyses demonstrate that CAHGCL consistently outperforms state-of-the-art baselines on three datasets. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.
Keyword ：
Adversarial machine learning Adversarial machine learning Contrastive Learning Contrastive Learning Federated learning Federated learning Knowledge graph Knowledge graph Self-supervised learning Self-supervised learning
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Li, Xinying , Wu, Ling , Guo, Kun . Community-Aware Heterogeneous Graph Contrastive Learning [C] . 2025 : 251-265 . |
| MLA | Li, Xinying et al. "Community-Aware Heterogeneous Graph Contrastive Learning" . (2025) : 251-265 . |
| APA | Li, Xinying , Wu, Ling , Guo, Kun . Community-Aware Heterogeneous Graph Contrastive Learning . (2025) : 251-265 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
The extrapolation task in the temporal knowledge graph has received increasing attention from scholars due to its wide range of practical application scenarios. At present, recurrent neural networks are currently widely used in temporal knowledge graph completion techniques. These networks are employed to depict the sequential pattern of entities and relations. However, as the sequence lengthens, some critical early information may become diluted. Prediction errors ensue in the completion task as a result. Furthermore, it is observed that existing temporal knowledge graph completion methods fail to account for the topological structure of relations, which leads to relation representations with essentially little distinction across different timestamps. In order to tackle the previously mentioned concern, our research introduces a Temporal Knowledge Graph Completion Method utilizing Sequence-Focus Patterns Representation Learning (SFP). This method contains two patterns: the Focus pattern and the Sequential pattern. In the SFP model, we developed a novel graph attention network called ConvGAT. This network efficiently distinguishes and extracts complex relation information, thereby enhancing the accuracy of entity representations that are aggregated in the Focus pattern and Sequential pattern. Furthermore we proposed RelGAT, a graph attention network that simulates the topological structure of relations. This enhances the precision of relation representations and facilitates the differentiation between relation embeddings generated at various timestamps in the Focus pattern. Utilizing a time-aware attention mechanism, the Focus pattern extracts vital information at particular timestamps in order to amplify the data that the Sequential pattern dilutes. On five distinct benchmark datasets, SFP significantly outperforms the baseline, according to a comprehensive series of experiments.
Keyword ：
Graph Attention Network Graph Attention Network Knowledge graph completion Knowledge graph completion Link prediction Link prediction Temporal knowledge graphs Temporal knowledge graphs
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Wang, Jingbin , Ke, Xifan , Zhang, Fuyuan et al. SFP: temporal knowledge graph completion based on sequence-focus patterns representation learning [J]. | APPLIED INTELLIGENCE , 2025 , 55 (6) . |
| MLA | Wang, Jingbin et al. "SFP: temporal knowledge graph completion based on sequence-focus patterns representation learning" . | APPLIED INTELLIGENCE 55 . 6 (2025) . |
| APA | Wang, Jingbin , Ke, Xifan , Zhang, Fuyuan , Wu, Yuwei , Zhang, Sirui , Guo, Kun . SFP: temporal knowledge graph completion based on sequence-focus patterns representation learning . | APPLIED INTELLIGENCE , 2025 , 55 (6) . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Although incremental methods are widely used in community detection, their error accumulation problem remains unresolved. Additionally, current methods typically identify events only after community detection has been completed for all time snapshots, lacking consideration of the impact of events on community structure during evolution. Therefore, this paper proposes a framework called Tracking dynamic community evolution based on Social Relevance and Strong Events(TranSiEnt). For the first time, TranSiEnt integrates evolution event identification with dynamic community updating, classifying evolution events into ordinary events and Strong Events based on the influence of the relevant communities. During dynamic community updating, TranSiEnt employs a path diffusion strategy to determine core nodes for community detection, establishing the initial community structure. Using an incremental approach, the framework expands the influence range of incremental nodes in communities experiencing Strong Events. It again conducts precise community detection on all affected nodes to reduce error accumulation, ultimately optimizing community partitioning. TranSiEnt was subjected to objective accuracy experiments on real and synthetic datasets, using modularity, NMI, and EMA as performance evaluation metrics. T-tests were used to verify the significance of the performance improvement of the TranSiEnt algorithm. The experimental results show that TranSiEnt performs better in dynamic community detection and evolution event tracking, significantly improving over existing methods.
Keyword ：
Dynamic community detection Dynamic community detection Social Relevance Social Relevance Strong Event Strong Event
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Wu, Ling , Xie, Xiaohua , Chen, Chengkai et al. Tracking dynamic community evolution based on Social Relevance and Strong Events [J]. | KNOWLEDGE AND INFORMATION SYSTEMS , 2025 . |
| MLA | Wu, Ling et al. "Tracking dynamic community evolution based on Social Relevance and Strong Events" . | KNOWLEDGE AND INFORMATION SYSTEMS (2025) . |
| APA | Wu, Ling , Xie, Xiaohua , Chen, Chengkai , Yang, Yingjie , Guo, Kun . Tracking dynamic community evolution based on Social Relevance and Strong Events . | KNOWLEDGE AND INFORMATION SYSTEMS , 2025 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Various datacenter network (DCN) load balancing schemes have been proposed in the past decade. Unfortunately, most of these solutions designed for lossy DCNs do not work well for Priority Flow Control (PFC) enabled lossless DCNs, primarily due to the reason that the individual congestion signals used in these solutions, e.g., link load, queue length, Round Trip Time (RTT) and Explicit Congestion Notification (ECN), may not be able to correctly or timely reflect the hop-by-hop PFC pausing. This paper first reveals the above problems via extensive experiments, and then based on the insights learned, we present Proteus, a PFC-aware load balancing scheme that is resilient to PFC pausing by exploring a combination of multi-level congestion signals. At its heart, Proteus leverages RTT-level signals (i.e., RTT and link utilization) to detect path status for initial routing decision, and exploits sub-RTT level signal (i.e., cumulative sojourn time) to reflect instantaneous PFC pausing and make timely rerouting choices based on the idea of better-late-than-never. We have implemented Proteus in the hardware programmable switch. Our testbed experiments as well as large-scale simulations show that Proteus can effectively handle PFC pausing under realistic workloads and achieve up to 35%, 31%, 28%, 22% and 46%, 42%, 34%, 29% better average FCT and 99(th) percentile FCT than CONGA, DRILL, Hermes and MP-RDMA, respectively.
Keyword ：
Computer science Computer science Datacenter Datacenter Delays Delays load balancing load balancing Load management Load management Load modeling Load modeling lossless networks lossless networks Receivers Receivers Switches Switches Transport protocols Transport protocols
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Hu, Jinbin , Zeng, Chaoliang , Wang, Zilong et al. Load Balancing With Multi-Level Signals for Lossless Datacenter Networks [J]. | IEEE-ACM TRANSACTIONS ON NETWORKING , 2024 , 32 (3) : 2736-2748 . |
| MLA | Hu, Jinbin et al. "Load Balancing With Multi-Level Signals for Lossless Datacenter Networks" . | IEEE-ACM TRANSACTIONS ON NETWORKING 32 . 3 (2024) : 2736-2748 . |
| APA | Hu, Jinbin , Zeng, Chaoliang , Wang, Zilong , Zhang, Junxue , Guo, Kun , Xu, Hong et al. Load Balancing With Multi-Level Signals for Lossless Datacenter Networks . | IEEE-ACM TRANSACTIONS ON NETWORKING , 2024 , 32 (3) , 2736-2748 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Abstract ：
Temporal knowledge graphs completion (TKGC) is a critical task that aims to forecast facts that will occur in future timestamps. It has attracted increasing research interest in recent years. Among the many approaches, reinforcement learning-based methods have gained attention due to their efficient performance and interpretability. However, these methods still face two challenges in the prediction task. First, a single policy network lacks the capability to capture the dynamic and static features of entities and relationships separately. Consequently, it fails to evaluate candidate actions comprehensively from multiple perspectives. Secondly, the composition of the action space is incomplete, often guiding the agent towards distant historical events and missing the answers in recent history. To address these challenges, this paper proposes a Temporal Knowledge Graph Completion Based on a Multi-Policy Network(MPNet). It constructs three policies from the aspects of static entity-relation, dynamic relationships, and dynamic entities, respectively, to evaluate candidate actions comprehensively. In addition, this paper creates a more diverse action space that guides the agent in investigating answers within historical subgraphs more effectively. The effectiveness of MPNet is validated through an extrapolation setting, and extensive experiments conducted on three benchmark datasets demonstrate the superior performance of MPNet compared to existing state-of-the-art methods.
Keyword ：
Knowledge graph completion Knowledge graph completion Link prediction Link prediction Reinforcement learning Reinforcement learning Temporal knowledge graphs Temporal knowledge graphs
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Wang, Jingbin , Wu, Renfei , Wu, Yuwei et al. MPNet: temporal knowledge graph completion based on a multi-policy network [J]. | APPLIED INTELLIGENCE , 2024 , 54 (3) : 2491-2507 . |
| MLA | Wang, Jingbin et al. "MPNet: temporal knowledge graph completion based on a multi-policy network" . | APPLIED INTELLIGENCE 54 . 3 (2024) : 2491-2507 . |
| APA | Wang, Jingbin , Wu, Renfei , Wu, Yuwei , Zhang, Fuyuan , Zhang, Sirui , Guo, Kun . MPNet: temporal knowledge graph completion based on a multi-policy network . | APPLIED INTELLIGENCE , 2024 , 54 (3) , 2491-2507 . |
| Export to | NoteExpress RIS BibTex |
Version ：
Export
| Results： |
Selected to |
| Format： |
