Exploiting higher-order patterns for community detection in attributed graphs

Hu, Lun; Pan, Xiangyu; Yan, Hong; Hu, Pengwei; He, Tiantian

doi:10.3233/ica-200645

Cited by 28 publications

(7 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, contrastive self-supervised learning is more challenging to converge; it requires more negative samples than pseudolabel-based methods, so designing a suitable contrastive selfsupervised method to make it suitable for incremental learning is what we plan to do shortly. In the long term, we may also try to introduce methods such as the automatic discovery of high-order patterns [38] to help incremental learning methods better alleviate catastrophic forgetting.…”

Section: Discussionmentioning

confidence: 99%

Reduce the Difficulty of Incremental Learning With Self-Supervised Learning

Guan

2021

IEEE Access

View full text Add to dashboard Cite

Incremental learning requires the model to learn new tasks without forgetting the learned tasks continuously. However, when the deep learning model learns new tasks, it will catastrophically forget the tasks it has learned before. Researchers have proposed methods to alleviate catastrophic forgetting; these methods only consider extracting features related to tasks learned before from learning samples, suppressing the model extracting features for unlearned tasks. As a result, when the model learns new tasks incrementally, the model needs to learn to extract the relevant features of the newly learned task quickly; this requires a significant change in the model's behavior of extracting features, which increases the learning difficulty. Therefore, the model is caught in the dilemma of reducing the learning rate to retain existing knowledge or increasing the learning rate to learn new knowledge quickly; we believe that introducing selfsupervised learning can alleviate this problem. Self-supervised learning methods learn to obtain supervised signals by customizing various pseudo-labels. Since the features extracted from the network model are based on self-supervised learning, independent of the specific subtasks of incremental learning, incorporating the supervised signal of self-supervised learning into the incremental learning process can help the model learn in incremental learning. The self-supervised learning process provides universal constraints, so that the model not only extracts features that are effective for the current learned task, but also allows the model to extract features that are suitable for other tasks that have not been learned, thereby allowing the network model parameters to iterate more smoothly and quickly to a point in the parameter space that satisfies the new and old tasks at the same time when learning a new task incrementally. The method of fusing selfsupervised and supervised learning signals significantly improve the adaptability of incremental learning without the need for additional labeled data. We verified that the introduction of self-supervised learning could significantly alleviate catastrophic forgetting in incremental learning algorithms based on sample replay by combining different databases and incremental learning model methods.INDEX TERMS incremental learning, self-supervised learning, deep learning

show abstract

Section: Discussionmentioning

confidence: 99%

Reduce the Difficulty of Incremental Learning With Self-Supervised Learning

Guan

2021

IEEE Access

View full text Add to dashboard Cite

show abstract

“…The third group is motif-oriented approaches [20][21][22][39][40][41][42]. Motifs lie between the microscopic proximity structure and mesoscopic community structure and help find communities that maintain building blocks in the network.…”

Section: Related Workmentioning

confidence: 99%

“…For attributed networks, they [41] first formulated an AHMotif adjacency matrix to encode node features and network topology from a higher-order perspective, and then utilized proximity-based methods to find communities. Hu et al [42] composed tensors to model higherorder patterns in terms of node features and network topology, and developed a novel algorithm to capture these patterns to find communities.…”

Section: Related Workmentioning

confidence: 99%

Triangle-oriented Community Detection considering Node Features and Network Topology

Gao¹,

Liang²,

Yuan³

et al. 2022

Preprint

View full text Add to dashboard Cite

The joint use of node features and network topology to detect communities is called community detection in attributed networks. Most of the existing work along this line has been carried out through objective function optimization and has proposed numerous approaches. However, they tend to focus only on lower-order details, i.e., capture node features and network topology from node and edge views, and purely seek a higher degree of optimization to guarantee the quality of the found communities, which exacerbates unbalanced communities and free-rider effect. To further clarify and reveal the intrinsic nature of networks, we conduct triangle-oriented community detection considering node features and network topology. Specifically, we first introduce a trianglebased quality metric to preserve higher-order details of node features and network topology, and then formulate so-called two-level constraints to encode lower-order details of node features and network topology. Finally, we develop a local search framework based on optimizing our objective function consisting of the proposed quality metric and two-level constraints to achieve both nonoverlapping and overlapping community detection in attributed networks. Extensive experiments demonstrate the effectiveness and efficiency of our framework and its potential in alleviating unbalanced communities and free-rider effect.

show abstract

“…However, there is an obvious disadvantage among these algorithms in that they only learn the embedding of the topology structure, without considering the attributes of nodes, which include chemical structures, targets, enzymes, and so on. As emphasized by [23,24], node attributes are essential for a precise analysis of complicated networks.…”

Section: Introductionmentioning

confidence: 99%

An Improved Graph Isomorphism Network for Accurate Prediction of Drug–Drug Interactions

Wang,

Su,

Zhao

et al. 2023

Mathematics

Self Cite

View full text Add to dashboard Cite

Drug–drug interaction (DDI) prediction is one of the essential tasks in drug development to ensure public health and patient safety. Drug combinations with potentially severe DDIs have been verified to threaten the safety of patients critically, and it is therefore of great significance to develop effective computational algorithms for identifying potential DDIs in clinical trials. By modeling DDIs with a graph structure, recent attempts have been made to solve the prediction problem of DDIs by using advanced graph representation learning techniques. Still, their representational capacity is limited by isomorphic structures that are frequently observed in DDI networks. To address this problem, we propose a novel algorithm called DDIGIN to predict DDIs by incorporating a graph isomorphism network (GIN) such that more discriminative representations of drugs can thus be learned for improved performance. Given a DDI network, DDIGIN first initializes the representations of drugs with Node2Vec according to the topological structure and then optimizes these representations by propagating and aggregating the first-order neighboring information in an injective way. By doing so, more powerful representations can thus be learned for drugs with isomorphic structures. Last, DDIGIN estimates the interaction probability for pairwise drugs by multiplying their representations in an end-to-end manner. Experimental results demonstrate that DDIGIN outperforms several state-of-the-art algorithms on the ogbl-ddi (Acc = 0.8518, AUC = 0.8594, and AUPR = 0.9402) and DDInter datasets (Acc = 0.9763, AUC = 0.9772, and AUPR = 0.9868). In addition, our case study indicates that incorporating GIN enhances the expressive power of drug representations for improved performance of DDI prediction.

show abstract

Exploiting higher-order patterns for community detection in attributed graphs

Cited by 28 publications

References 36 publications

Reduce the Difficulty of Incremental Learning With Self-Supervised Learning

Reduce the Difficulty of Incremental Learning With Self-Supervised Learning

Triangle-oriented Community Detection considering Node Features and Network Topology

An Improved Graph Isomorphism Network for Accurate Prediction of Drug–Drug Interactions

Contact Info

Product

Resources

About