Detecting High-Potential SMEs with Heterogeneous Graph Neural Networks

TL;DR

This paper presents SME-HGT, a heterogeneous graph transformer model that predicts high-potential SMEs using public data, outperforming baselines and aiding policymakers and investors in early-stage decision-making.

Contribution

Introduction of SME-HGT, a novel heterogeneous graph neural network framework leveraging relational data to predict SME success, with improved accuracy over existing methods.

Findings

SME-HGT achieves an AUPRC of 0.621, outperforming baselines.

At screening depth 100, SME-HGT attains 89.6% precision.

Relational structure among entities provides meaningful signals for SME potential.

Abstract

Small and Medium Enterprises (SMEs) constitute 99.9% of U.S. businesses and generate 44% of economic activity, yet systematically identifying high-potential SMEs remains an open challenge. We introduce SME-HGT, a Heterogeneous Graph Transformer framework that predicts which SBIR Phase I awardees will advance to Phase II funding using exclusively public data. We construct a heterogeneous graph with 32,268 company nodes, 124 research topic nodes, and 13 government agency nodes connected by approximately 99,000 edges across three semantic relation types. SME-HGT achieves an AUPRC of 0.621 0.003 on a temporally-split test set, outperforming an MLP baseline (0.590 0.002) and R-GCN (0.608 0.013) across five random seeds. At a screening depth of 100 companies, SME-HGT attains 89.6% precision with a 2.14 lift over random selection. Our temporal evaluation protocol prevents information leakage, and our reliance on public data ensures reproducibility. These results demonstrate that relational structure among firms, research topics, and funding agencies provides meaningful signal for SME potential assessment, with implications for policymakers and early-stage investors.