Reconnecting Fragmented Citation Networks with Semantic Augmentation
Authors: Vu Thi Huong, Annika Buchholz, Imene Khebouri, Thorsten Koch, Tim Kunt, Wolfgang Peters-Kottig, Tomasz Stompor, Janina Zittel
Organizations: Digital Data and Information for Society, Science, and Culture, Zuse Institute Berlin, Takustr. 7, 14195 Berlin, Germany · Institute of Mathematics, Vietnam Academy of Science and Technology, 10072 Hanoi, Vietnam · Software and Algorithms for Discrete Optimization, Technische Universität Berlin, Straße des 17. Juni 135, 10623 Berlin, Germany · Applied Optimization, Zuse Institute Berlin, Takustr. 7, 14195 Berlin, Germany · Kooperativer Bibliotheksverbund Berlin-Brandenburg (KOBV), Zuse Institute Berlin, Takustr. 7, 14195 Berlin, Germany
Citation graphs are fundamental tools for modeling scientific structure, but are often fragmented due to missing citations of scientifically connected articles. To address this issue, we propose a computationally efficient hybrid framework integrating citation topology with large language model (LLM)-based text similarity. Using 662,369 Web of Science publications in Mathematics and Operations Research & Management Science, we augment the original graph by adding semantic edges from small, disconnected components and weighting existing citations according to textual similarity. Semantic augmentation substantially reduces fragmentation while preserving disciplinary homogeneity. Compared to embedding-only clustering, cluster detection on augmented graphs using the Leiden algorithm retains structural interpretability while offering multi-scale organization. The method scales efficiently to large datasets and offers a practical strategy for strengthening citation-based indicators without collapsing disciplinary boundaries.