Abstract:As data in many disciplines continues to grow at a frantic pace, a more heterogeneous set of information sources both can and need to be connected in innovative ways to improve information search and recommendation systems. Using information retrieval (IR) as an example, classical search techniques are based on limited information, including query terms, the content of documents, and basic user characteristics. However, recent studies have started to investigate additional information resources, including semantic web data, knowledge graphs, linguistic features, user email account content, social networks, user behavior, the search context, and other sources. Intuitively, a heterogeneous graph integrating all these information sources could enhance search performance. However, the complexity of such graphs challenge most existing graph mining algorithms. In this study, we propose a novel approach for a personalized random walk over a complex heterogeneous graph; which we refer to as Personalized Graph Navigation (PGN). Unlike earlier expert-guided random walk approaches, by using an EM framework, PGN estimates the personalized usefulness probability distribution for each edge type (the latent variable), which can be used as a random walk navigation profile for a user on a heterogeneous graph. While PGN can cope with information retrieval/recommendation problems at a low cost, this method also transforms a complex heterogeneous graph into a homogeneous graph, allowing existing homogeneous graph mining algorithms to be applied to a heterogeneous graph.

Feature extraction using randomwalks

Commute time guided transformation for feature extraction

Feature Extraction Via Multi-View Non-Negative Matrix Factorization with Local Graph Regularization.

Personalized Navigation and Random Walk on a Complex Heterogeneous Graph

Feature extraction using graph discriminant embedding

Feature extraction based on graph discriminant embedding and its applications to face recognition

A two-view based multilayer feature graph for robot navigation

Non-local Gait Feature Extraction and Human Identification

The Hidden Flow Structure and Metric Space of Network Embedding Algorithms Based on Random Walks

Estimation of Graph Features Based on Random Walks Using Neighbors' Properties

General Moving Objects Recognition Method Based on Graph Embedding Dimension Reduction Algorithm

Estimating and Sampling Graphs with Multidimensional Random Walks

Multi-label Feature Extraction With Distance-Based Graph Attention Network.

A Broader Picture of Random-walk Based Graph Embedding

Random Walk on Multiple Networks

From random walks to distances on unweighted graphs

General Graph Random Features

Image Feature Extraction via Graph Embedding Regularized Projective Non-negative Matrix Factorization.

Variational Graph Embedding for Globally and Locally Consistent Feature Extraction

Random Walk Conformer: Learning Graph Representation from Long and Short Range

Detecting Community Structure for Undirected Big Graphs Based on Random Walks