Modeling homophily and stochastic equivalence in symmetric relational data

Abstract

This article discusses a latent variable model for inference and prediction of symmetric relational data. The model, based on the idea of the eigenvalue decomposition, represents the relationship between two nodes as the weighted inner-product of node-specific vectors of latent characteristics. This ``eigenmodel'' generalizes other popular latent variable models, such as latent class and distance models: It is shown mathematically that any latent class or distance model has a representation as an eigenmodel, but not vice-versa. The practical implications of this are examined in the context of three real datasets, for which the eigenmodel has as good or better out-of-sample predictive performance than the other two models.

arXiv comment12 pages, 4 figures, 1 table

Cited by in corpus (72)

• A Review of Relational Machine Learning for Knowledge Graphs
• Fast community detection by SCORE
• Consistency of community detection in networks under degree-corrected stochastic block models
• Pseudo-likelihood methods for community detection in large sparse networks
• Stochastic blockmodels with growing number of classes
• Rate-optimal graphon estimation
• Community extraction for social networks
• Evaluating Overfit and Underfit in Models of Network Community Structure
• Multilinear tensor regression for longitudinal relational data
• The method of moments and degree distributions for network models
• Stochastic blockmodel approximation of a graphon: Theory and consistent estimation
• Infinite Edge Partition Models for Overlapping Community Detection and Link Prediction
• Exponential-Family Models of Random Graphs: Inference in Finite-, Super-, and Infinite Population Scenarios
• Discovering Latent Network Structure in Point Process Data
• Max-Margin Nonparametric Latent Feature Models for Link Prediction
• Bayesian Inference and Testing of Group Differences in Brain Networks
• A Consistent Histogram Estimator for Exchangeable Graph Models
• Mixed-Membership of Experts Stochastic Blockmodel
• Max-Margin Nonparametric Latent Feature Models for Link Prediction
• Bayesian latent structure discovery from multi-neuron recordings
• Locally Adaptive Dynamic Networks
• Bayesian Network--Response Regression
• Dyadic data analysis with amen
• Estimating network edge probabilities by neighborhood smoothing
• Social Network Mediation Analysis: a Latent Space Approach
• Network cross-validation by edge sampling
• Sparse matrix-variate Gaussian process blockmodels for network modeling
• Confidence sets for network structure
• The Multiple Random Dot Product Graph Model
• Network Representation Using Graph Root Distributions
• Community Detection in Networks with Node Features
• Exploration of Large Networks with Covariates via Fast and Universal Latent Space Model Fitting
• Network Cross-Validation for Determining the Number of Communities in Network Data
• Inferring Influence Networks from Longitudinal Bipartite Relational Data
• Generalized linear models with low rank effects for network data
• Comparison of Cross-Validation Methods for Stochastic Block Models
• Link prediction for egocentrically sampled networks
• Pairwise Covariates-adjusted Block Model for Community Detection
• Hierarchical community detection by recursive partitioning
• Inferential Approaches for Network Analyses: AMEN for Latent Factor Models
• An iterative step-function estimator for graphons
• Goodness of fit of logistic models for random graphs
• A Latent Parameter Node-Centric Model for Spatial Networks
• Dropout Training of Matrix Factorization and Autoencoder for Link Prediction in Sparse Graphs
• Spectral clustering via adaptive layer aggregation for multi-layer networks
• Community detection for interaction networks
• Topics in social network analysis and network science
• Multiresolution Tensor Decomposition for Multiple Spatial Passing Networks
• Bayesian analysis of matrix data with rstiefel
• The Importance of Being Correlated: Implications of Dependence in Joint Spectral Inference across Multiple Networks
• A Review of Latent Space Models for Social Networks
• Common and Individual Structure of Brain Networks
• Simultaneous Dimensionality and Complexity Model Selection for Spectral Graph Clustering
• Discriminative Nonparametric Latent Feature Relational Models with Data Augmentation
• Factor Models for High-Dimensional Dynamic Networks: with Application to International Trade Flow Time Series 1981-2015
• Community detection in sparse latent space models
• Ranking relations using analogies in biological and information networks
• Identifiability and consistency of network inference using the hub model and variants
• Manifold structure in graph embeddings
• Network Mediation Analysis Using Model-based Eigenvalue Decomposition
• Testing for nodal dependence in relational data matrices
• Multiplicative Coevolution Regression Models for Longitudinal Networks and Nodal Attributes
• A Latent Space Model for Multilayer Network Data
• Selection of Exponential-Family Random Graph Models via Held-Out Predictive Evaluation (HOPE)
• Discriminative Relational Topic Models
• Exact Inference with Latent Variables in an Arbitrary Domain
• A Dynamic Additive and Multiplicative Effects Model with Application to the United Nations Voting Behaviors
• Outliers Detection in Networks with Missing Links
• A de Finetti-type representation of joint hierarchically exchangeable arrays on directed acyclic graphs
• An Eigenmodel for Dynamic Multilayer Networks
• Network estimation via graphon with node features
• Process convolution approaches for modeling interacting trajectories