The hidden structure of innovation networks

TL;DR

This study analyzes the mesoscopic structure of innovation networks in AI, biotech, and semiconductors using patent data, revealing dense inventor clusters and hierarchical organization networks linked to innovation impact.

Contribution

It introduces a Bayesian inference-based method to better capture the hierarchical meso-structures in innovation networks compared to traditional modularity maximization.

Findings

Inventor networks are denser and more clustered than organization networks.

Organization networks exhibit clear hierarchical role-based structures.

Innovation impact is concentrated in a few key clusters, indicating inequality in technological influence.

Abstract

Innovation emerges from complex collaboration patterns - among inventors, firms, or institutions. However, not much is known about the overall mesoscopic structure around which inventive activity self-organizes. Here, we tackle this problem by employing patent data to analyze both individual (co-inventorship) and organization (co-ownership) networks in three strategic domains (artificial intelligence, biotechnology and semiconductors). We characterize the mesoscale structure (in terms of clusters) of each domain by comparing two alternative methods: a standard baseline - modularity maximization - and one based on the minimization of the Bayesian Information Criterion, within the Stochastic Block Model and its degree-corrected variant. We find that, across sectors, inventor networks are denser and more clustered than organization ones - consistent with the presence of small recurrent teams embedded into broader institutional hierarchies - whereas organization networks have neater hierarchical role-based structures, with few bridging firms coordinating the most peripheral ones. We also find that the discovered meso-structures are connected to innovation output. In particular, Lorenz curves of forward citations show a pervasive inequality in technological influence: across sectors and methods, both inventor (especially) and organization networks consistently show high levels of concentration of citations in a few of the discovered clusters. Our results demonstrate that the baseline modularity-based method may not be capable of fully capturing the way collaborations drive the spreading of inventive impact across technological domains. This is due to the presence of local hierarchies that call for more refined tools based on Bayesian inference.