Abstract:By using edges to represent pairwise interactions, networks provide a powerful formalism to model complex systems. As a fundamental problem in network science, link prediction has been widely studied. But in the real-world systems, interactions often take place among more than two nodes at once, such as research collaborations among more than two authors and social activities involving multiple individuals. To capture these interactions and describe the real-world systems more accurately, higher-order networks have been studied over the past years. And as a challenging problem in higher-order networks, higher-order link problem has also gained more and more attention. Recently, researchers have found that local structures tend to play an important role in predicting higher-order link existence. However, existing methods based on local information have the problems of limited applying scenarios. To find a more general solution and make better use of local information, a reasonable way should be automatically learning information from enclosing subgraphs extracted around target links. In this paper, we propose a higher-order link prediction framework to learn graph structure features from local enclosing subgraphs using Message Passing Simplicial Networks. To evaluate the performance of our framework, we conduct experiments on eight empirical networks and the results show that it is a superb method with unprecedented performance on a wide range of datasets.

Higher-order Link Prediction Based on Message Passing Simplicial Networks

Higher-order link prediction via local information

Pure Message Passing Can Estimate Common Neighbor for Link Prediction

Simplicial closure and higher-order link prediction

Effective Model Integration Algorithm for Improving Link and Sign Prediction in Complex Networks

A Comprehensive Framework for Link Prediction in Multiplex Networks

Link Prediction Based on Local Structure and Node Information along Local Paths.

Link Prediction Based on the Mutual Information with High-Order Clustering Structure of Nodes in Complex Networks

Effective Link Prediction Based on Community Relationship Strength.

A Novel Complex Network Link Prediction Framework Via Combining Mutual Information with Local Naive Bayes

Link Prediction In Multi-Relational Collaboration Networks

Community and Local Information Preserved Link Prediction in Complex Networks.

Link Prediction in Complex Networks: a Local Naïve Bayes Model

Bi-scale link prediction on networks

Accurate and Fast Link Prediction in Complex Networks

Link Prediction Via Layer Relevance of Multiplex Networks

Interlayer Link Prediction Based on Multiple Network Structural Attributes

Link Prediction in Complex Networks Based on Communication Capacity and Local Paths

Link Prediction with Untrained Message Passing Layers

Link Prediction Based on Hyper-Substructure Network

Link prediction in heterogeneous information networks: An improved deep graph convolution approach