Abstract:Community search aims to provide efficient solutions for searching high-quality communities via given sample nodes from network. Much research effort has devoted to mining a single community based on the assumption of sample nodes are from the same community. Despite their effectiveness, the following two insights are often neglected. First, complex high-order structural relationship and attribute provide auxiliary information to represent nodes and offer meaningful information to compensate the incomplete and missing information of the network, it benefits to optimal results. Second, user usually assumes that sample nodes come from the same community without any prior knowledge. This stringent assumption limits the flexibility of algorithm in many real-world scenarios. To this end, we propose a novel multi-community search method in attributed networks that is capable of effectively searching multi-communities where sample nodes locate. Specifically, we appoint complex structural information as internal attributes which explicitly encode node's interaction and combine it with node attributes. In order to better capture association between nodes and attributes, we construct a node-attribute graph and similarity enhanced random walk is performed based on it. The similarity enhanced random walk is developed to reinforce the walking path of each sample node so that they can better distinguish and capture the community structure for sample nodes. The multi-communities with densely connected and similar attributes can be found by parallel conductance. Extensive experimental results on both synthetic and real-world graphs verify the effectiveness and efficiency of the proposed method, and show its superiority over many state-of-art approaches.

Discovering Personalized Characteristic Communities in Attributed Graphs

Effective Influential Community Search on Attributed Graph

Top-r Keyword-Based Community Search in Attributed Graphs.

Effective Community Search on Large Attributed Bipartite Graphs.

Effective Community Search on Large Attributed Bipartite Graphs

Community Detection in Attributed Graphs: an Embedding Approach

An Effective Method for Community Search in Large Directed Attributed Graphs.

Exploiting higher-order patterns for community detection in attributed graphs

Community Search for Multiple Nodes on Attribute Graphs

Influential Attribute Community Search

Personalized Web Page Ranking Based Graph Convolutional Network for Community Detection in Attribute Networks

Attributed Community Search in Dynamic Networks.

QD-GCN: Query-Driven Graph Convolutional Networks for Attributed Community Search

Attributed Multi-Query Community Search Via Random Walk Similarity.

Codar: Revealing The Generalized Procedure & Recommending Algorithms Of Community Detection

Efficient Search of the Most Cohesive Co-located Community in Attributed Networks

Attribute-Driven Community Search.

Personalized Top-n Influential Community Search over Large Social Networks

Topic-aware Most Influential Community Search in Social Networks

CDIA: A Feasible Community Detection Algorithm Based on Influential Nodes in Complex Networks

Top-r Influential Communities Identification over Attributed Networks