Paper Number
ECIS2025-1734
Paper Type
SP
Abstract
Multi-platform ecosystems form intricate network structures that create challenges for identifying complementarities. Existing analytical frameworks primarily address single-platform environments, failing to capture cross-platform interactions, indirect value relationships, and higher-order complementarities that require multiple components to interact simultaneously before value materializes. This study presents a novel graph neural network (GNN) approach for identifying complementarities in multi-platform ecosystems, representing platforms and complements as nodes and their relation-ships as edges. The empirical validation with the Alexa Skills ecosystem demonstrates a strong dis-criminatory ability across multiple performance metrics while providing interpretable insights through attention mechanisms. Our method advances platform theory by extending complementarity analysis to networked ecosystems and offers platform owners and complementors a scalable technique for discovering value opportunities across platform boundaries.
Recommended Citation
Schmidt, Rainer; Alt, Rainer; and ZIMMERMANN, ALFRED, "PREDICTING COMPLEMENTARITIES IN MULTI-PLATFORM ECOSYSTEMS: A GRAPH NEURAL NETWORK APPROACH" (2025). ECIS 2025 Proceedings. 2.
https://aisel.aisnet.org/ecis2025/gov_platform/gov_platform/2
PREDICTING COMPLEMENTARITIES IN MULTI-PLATFORM ECOSYSTEMS: A GRAPH NEURAL NETWORK APPROACH
Multi-platform ecosystems form intricate network structures that create challenges for identifying complementarities. Existing analytical frameworks primarily address single-platform environments, failing to capture cross-platform interactions, indirect value relationships, and higher-order complementarities that require multiple components to interact simultaneously before value materializes. This study presents a novel graph neural network (GNN) approach for identifying complementarities in multi-platform ecosystems, representing platforms and complements as nodes and their relation-ships as edges. The empirical validation with the Alexa Skills ecosystem demonstrates a strong dis-criminatory ability across multiple performance metrics while providing interpretable insights through attention mechanisms. Our method advances platform theory by extending complementarity analysis to networked ecosystems and offers platform owners and complementors a scalable technique for discovering value opportunities across platform boundaries.
When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.