GraphGPT-Patent: Time-Aware Graph Foundation Modeling on Semantic Similarity Document Graphs for Grant-Time Economic Impact Prediction
Applied Sciences · 12.03.2026 · 16 (6) · 2737
DOI: 10.3390/app16062737 ↗Kurz gefasst
Fachartikel in der Zeitschrift Applied Sciences, online erschienen am 12.03.2026. Verfasst von Tianhui Fang und 3 weiteren Autor:innen. Die Arbeit nennt 30 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Predicting the future impact of technical economic documents at release time is challenging due to delayed supervision signals, long-tailed label distributions, and time- and domain-dependent shifts in language and topics. Moreover, similarity graphs derived from text embeddings can be noisy due to boilerplate and evolve under temporal drift, making robustness and leakage-free evaluation essential. We formulate grant-time patent impact prediction as a node classification and within-domain ranking problem on a large-scale semantic similarity document graph built from patent text embeddings, avoiding any future citation leakage. The document graph is constructed via ANN Top-K retrieval and similarity thresholding, enabling scalable and reproducible sparsification on hundreds of thousands of nodes. We propose GraphGPT-Patent, which adapts a reversible graph-to-sequence foundation backbone to local subgraphs extracted from the similarity network. The model incorporates time- and domain-conditioned edge reliability to suppress drift-induced and template-driven pseudo-similarity, and optimizes a joint objective coupling high-impact classification with ranking consistency within comparable groups. Experiments on USPTO granted patents (2000–2022) across three high-volume CPC domains and three evaluation horizons show consistent gains over text-only and GNN baselines, achieving up to 0.94 recall for the positive class and improved macro-average recall across nine settings. Temporal shift analyses further quantify the effect of training-data freshness, while explanation subgraphs provide auditable structural evidence of model decisions. The proposed framework offers an effective graph-based learning pipeline for scalable impact prediction and downstream triage under strict information constraints.
Zitieren
APA
Fang, T., Si, J., Ye, C., & Shi, H. (2026). GraphGPT-Patent: Time-Aware Graph Foundation Modeling on Semantic Similarity Document Graphs for Grant-Time Economic Impact Prediction. Applied Sciences, 16(6), 2737. https://doi.org/10.3390/app16062737
Vancouver
Fang T, Si J, Ye C, Shi H. GraphGPT-Patent: Time-Aware Graph Foundation Modeling on Semantic Similarity Document Graphs for Grant-Time Economic Impact Prediction. Applied Sciences. 2026;16(6):2737. doi:10.3390/app16062737
BibTeX
@article{Fang2026,
author = {Fang, Tianhui and Si, Junru and Ye, Chi and Shi, Hailong},
title = {GraphGPT-Patent: Time-Aware Graph Foundation Modeling on Semantic Similarity Document Graphs for Grant-Time Economic Impact Prediction},
journal = {Applied Sciences},
year = {2026},
volume = {16},
number = {6},
pages = {2737},
doi = {10.3390/app16062737}
}
EUVIMED