Abstract:Searching for a community containing a query node in an online social network enjoys wide applications like recommendation, team organization, etc. When applied to real-life networks, the existing approaches face two major limitations. First, they usually take two steps, i.e., crawling a large part of the network first and then finding the community next, but the entire network is usually too big and most of the data are not interesting to end users. Second, the existing methods utilize hand-crafted rules to measure community membership, while it is very difficult to define effective rules as the communities are flexible for different query nodes. This paper proposes an interactive community search method based on graph neural network (shortened by ICS-GNN(+)) to locate the target community over a subgraph collected on the fly from an online network iteratively. In each iteration, we first build a candidate subgraph around the query node and labeled nodes. We then train a node classification model using GNN to determine whether every node belongs to the target community, which captures similarities between nodes by combining content and structural features seamlessly and flexibly under the guide of users' labeling. Based on the probabilities inferred from the trained GNN, we introduce a k-sized Maximum-GNN-scores (shortened by kMG) community to describe the target community and design a method to locate the kMG community which will be evaluated by end users to acquire more feedback. Besides, various optimization strategies are proposed including an adaptive method to maintain the subgraph during iterations, combining ranking loss into the GNN model, generating node embedding enhanced by pseudo-labels from node clusters in the subgraph, and a greedy community searching method with benefit computed globally. We conduct the experiments on both offline and online real-life datasets, and demonstrate that ICS-GNN(+) can produce effective communities with low overhead in communication, computation, and user labeling.

Efficient Learning-based Community-Preserving Graph Generation

Detecting User Community in Sparse Domain via Cross-Graph Pairwise Learning

CommunityGAN: Community Detection with Generative Adversarial Nets

Semi-supervised community detection method based on generative adversarial networks

GraphGAN: Graph Representation Learning with Generative Adversarial Nets

Deep Graph-Convolutional Generative Adversarial Network for Semi-Supervised Learning on Graphs

Dynamic and Textual Graph Generation Via Large-Scale LLM-based Agent Simulation

DyGCN: Efficient Dynamic Graph Embedding With Graph Convolutional Network

Graph Generative Model for Benchmarking Graph Neural Networks

Adaptive Sampling Towards Fast Graph Representation Learning

Community enhanced graph convolutional networks

Geodesic Graph Neural Network for Efficient Graph Representation Learning

Xgcn: an Extreme Graph Convolutional Network for Large-scale Social Link Prediction

Scalable Deep Generative Modeling for Sparse Graphs

ICS-GNN+: Lightweight Interactive Community Search Via Graph Neural Network

Community hiding using a graph autoencoder

CGGM: A conditional graph generation model with adaptive sparsity for node anomaly detection in IoT networks

Dyngraphgan: Dynamic Graph Embedding Via Generative Adversarial Networks

FastSGG: Efficient Social Graph Generation Using a Degree Distribution Generation Model

Accurate, Efficient and Scalable Graph Embedding

Efficient and Scalable Graph Generation through Iterative Local Expansion