Including Node Textual Metadata in Laplacian-constrained Gaussian Graphical Models

TL;DR

This paper introduces a novel method for learning Gaussian Graphical Models that incorporates node textual metadata, improving graph clustering by jointly leveraging signals and metadata with an efficient optimization algorithm.

Contribution

It proposes a Laplacian-constrained GGM approach that fuses node signals and textual metadata, along with a majorization-minimization algorithm for efficient optimization.

Findings

Significantly improves graph clustering performance on real-world data.

Demonstrates the benefit of combining node signals with textual metadata.

Outperforms state-of-the-art methods using only signals or metadata.

Abstract

This paper addresses graph learning in Gaussian Graphical Models (GGMs). In this context, data matrices often come with auxiliary metadata (e.g., textual descriptions associated with each node) that is usually ignored in traditional graph estimation processes. To fill this gap, we propose a graph learning approach based on Laplacian-constrained GGMs that jointly leverages the node signals and such metadata. The resulting formulation yields an optimization problem, for which we develop an efficient majorization-minimization (MM) algorithm with closed-form updates at each iteration. Experimental results on a real-world financial dataset demonstrate that the proposed method significantly improves graph clustering performance compared to state-of-the-art approaches that use either signals or metadata alone, thus illustrating the interest of fusing both sources of information.