Abstract:Recent years have witnessed a surge of interest in machine learning on graphs and networks with applications ranging from vehicular network design to IoT traffic management to social network recommendations. Supervised machine learning tasks in networks such as node classification and link prediction require us to perform feature engineering that is known and agreed to be the key to success in applied machine learning. Research efforts dedicated to representation learning, especially representation learning using deep learning, has shown us ways to automatically learn relevant features from vast amounts of potentially noisy, raw data. However, most of the methods are not adequate to handle heterogeneous information networks which pretty much represents most real-world data today. The methods cannot preserve the structure and semantic of multiple types of nodes and links well enough, capture higher-order heterogeneous connectivity patterns, and ensure coverage of nodes for which representations are generated. We propose a novel efficient algorithm, motif2vec that learns node representations or embeddings for heterogeneous networks. Specifically, we leverage higher-order, recurring, and statistically significant network connectivity patterns in the form of motifs to transform the original graph to motif graph(s), conduct biased random walk to efficiently explore higher order neighborhoods, and then employ heterogeneous skip-gram model to generate the embeddings. Unlike previous efforts that uses different graph meta-structures to guide the random walk, we use graph motifs to transform the original network and preserve the heterogeneity. We evaluate the proposed algorithm on multiple real-world networks from diverse domains and against existing state-of-the-art methods on multi-class node classification and link prediction tasks, and demonstrate its consistent superiority over prior work.

How the four-nodes motifs work in heterogeneous node representation?

motif2vec: Motif Aware Node Representation Learning for Heterogeneous Networks

User-Guided Clustering in Heterogeneous Information Networks via Motif-Based Comprehensive Transcription

Unifying Homophily and Heterophily Network Transformation via Motifs

Higher-Order Clustering in Heterogeneous Information Networks.

Heterogeneous Graph Attention Network with Motif Clique

Heterogeneous Deep Graph Infomax

MMNR - A Network Representation Framework Based on Multi-View Motif Fusion.

Learning Node Representation Via Motif Coarsening

Self-supervised Nodes-Hyperedges Embedding for Heterogeneous Information Network Learning

Deep Learning-based Heterogeneous Information Network Representation:A Survey

Improving Sign Prediction of Network Embedding by Adding Motif Features

Exploring Multiple Hypergraphs for Heterogeneous Graph Neural Networks

Key Nodes Cluster Augmented Embedding for Heterogeneous Information Networks.

LIHAN: A Lattice-Guided Incomplete Heterogeneous Information Network Embedding Model for Node Classification

MBRep: Motif-based representation learning in heterogeneous networks

CN-Motifs Perceptive Graph Neural Networks

Efficient Maximal Motif-Clique Enumeration over Large Heterogeneous Information Networks

HINChip: Heterogeneous Information Network Representation with Community Hierarchy Preserving

Motif-based community detection in heterogeneous multilayer networks

Molecular Representation Learning via Heterogeneous Motif Graph Neural Networks