Düğüm Önemine İlişkin Semantik ve Yapısal Perspektifler: Klasik ve LLM Yaklaşımının Birleştirilmesi

Year 2026, Volume: 14 Issue: 1, 225 - 239, 21.01.2026

Levent Sabah , Zeynep Bozdoğan , Resul Kara

Abstract

Karmaşık ağlardaki düğüm önemine ilişkin araştırmalar, klasik merkezilik ölçütleri, entropi tabanlı ölçümler ve çeşitli makine öğrenme modelleri de dahil olmak üzere geniş bir yelpazede yaklaşımlardan yararlanmaya devam etmektedir. Önceki çalışmalar, entropi odaklı ve öğrenme tabanlı tekniklerin, bilgi yayılımı veya sağlamlık analizi gibi görevler için daha yüksek doğruluk ve daha hızlı yakınsama sağladığını sıklıkla bildirmektedir; ancak bu sonuçlar veri kümesine ve her yöntemin varsayımlarına bağlı olarak değişebilir. Daha az araştırılan konu ise, bu tekniklerin, yapısal verilerde doğrudan görünmeyen düğümlerin bağlamsal veya anlamsal özelliklerini ne ölçüde hesaba kattığıdır. Bu çalışmada, yerleşik merkezilik ölçütlerini büyük bir dil modeli (LLM) tarafından üretilen değerlendirmelerle birleştirerek bu boşluğu gidermek için hibrit bir çerçeve sunulmaktadır. Yaklaşım, Zachary'nin Karate Kulübü, Krackhardt Uçurtma Grafiği ve Caz Müzisyenleri Ağı gibi iyi bilinen birkaç kıyaslama ağında incelenmiş ve elde edilen sıralamalar, düğümler arasında daha net bir ayrım ve birçok durumda, tamamen yapısal temellerden daha yorumlanabilir kalıplar göstermiştir. Bu bulguları daha geniş literatür içinde konumlandırmak için, sonuçlar, bileşik veya entropi tabanlı merkezilik formülasyonlarını kullanan üç önceki çalışma ile karşılaştırıldı. Ölçeklenebilirliği araştırmak için, SNAP deposundan üç büyük işbirliği ağı da değerlendirildi. Genel kanıtlar, özellikle yapısal bilginin tek başına eksik bir tablo sunduğu durumlarda, LLM destekli puanlamanın karmaşık ağ analizi için tamamlayıcı bir bakış açısı sunabileceğini göstermektedir.

Keywords

Karmaşık ağ , Merkezilik ölçütü , Büyük dil modeli , Düğüm önemi

Semantic and Structural Perspectives on Node Importance: A Combined Classical–LLM Approach

Abstract

Research on node importance in complex networks continues to draw on a wide range of approaches, including classical centrality metrics, entropy-based measures, and various machine-learning models. Previous studies often reports that entropy-driven and learning-based techniques yield higher accuracy and faster convergence for tasks such as information spreading or robustness analysis, although these results can vary depending on the dataset and the assumptions of each method. What remains less explored is the extent to which these techniques account for the contextual or semantic characteristics of nodes, which are not directly visible in structural data. In this study, a hybrid framework is introduced to address this gap by combining established centrality metrics with evaluations produced by a large language model (LLM). The approach was examined on several well-known benchmark networks—Zachary’s Karate Club, the Krackhardt Kite Graph, and the Jazz Musicians Network—and the resulting rankings showed clearer separation among nodes and, in many cases, more interpretable patterns than the purely structural baselines. To situate these findings within the broader literature, results were compared with three prior studies that employ composite or entropy-based centrality formulations. To explore scalability, three larger collaboration networks from the SNAP repository were also evaluated. The overall evidence suggests that LLM-supported scoring may offer a complementary perspective for complex network analysis, especially when structural information alone provides an incomplete picture.

Keywords

Complex network , Centrality measure , Large language model , Node importance

Ethical Statement

This study does not involve human or animal participants. All procedures followed scientific and ethical principles, and all referenced studies are appropriately cited.

Supporting Institution

This research received no external funding.

Thanks

No acknowledgment statement is included in the manuscript.

References

• Ait Rai, K., Machkour, M., & Antari, J. (2023). Influential nodes identification in complex networks: a comprehensive literature review. Beni-Suef University Journal of Basic and Applied Sciences, 12(1), Article 18. https://doi.org/10.1186/s43088-023-00357-w
• Chaudhary, L., & Singh, B. (2023). Gumbel-SoftMax based graph convolution network approach for community detection. International Journal of Information Technology, 15(6), 3063–3070. https://doi.org/10.1007/s41870-023-01347-y
• Chen, D., Lü, L., Shang, M.-S., Zhang, Y.-C., & Zhou, T. (2012). Identifying influential nodes in complex networks. Physica A: Statistical Mechanics and Its Applications, 391(4), 1777–1787. https://doi.org/10.1016/j.physa.2011.09.017
• Chen, L., Xi, Y., Dong, L., Zhao, M., Li, C., Liu, X., & Cui, X. (2024). Identifying influential nodes in complex networks via Transformer. Information Processing & Management, 61(5), Article 103775. https://doi.org/10.1016/j.ipm.2024.103775
• DeepSeek-AI, Bi, X., Chen, D., Chen, G., Chen, S., Dai, D., Deng, C., Ding, H., Dong, K., Du, Q., Fu, Z., Gao, H., Gao, K., Gao, W., Ge, R., Guan, K., Guo, D., Guo, J., Guangbo, H., … Zou, Y. (2024). DeepSeek LLM: Scaling open-source language models with long-termism. arXiv. https://doi.org/10.48550/arXiv.2401.02954
• Dong, S., & Zhou, W. (2021). Improved influential nodes identification in complex networks. Journal of Intelligent & Fuzzy Systems, 41(6), 6263–6271. https://doi.org/10.3233/JIFS-202943
• Fei, L., & Deng, Y. (2017). A new method to identify influential nodes based on relative entropy. Chaos, Solitons & Fractals, 104, 257–267. https://doi.org/10.1016/j.chaos.2017.08.010
• Freeman, L. C. (1978). Centrality in social networks: Conceptual clarification. Social Networks, 1(3), 215–239. https://doi.org/10.1016/0378-8733(78)90021-7
• Gleiser, P. M., & Danon, L. (2003). Community structure in jazz. Advances in Complex Systems, 6(4), 565–573. https://doi.org/10.1142/S0219525903001067
• Guo, C., Yang, L., Chen, X., Chen, D., Gao, H., & Ma, J. (2020). Influential nodes identification in complex networks via information entropy. Entropy, 22(2), Article 242. https://doi.org/10.3390/e22020242
• Huang, Z., Tang, Y., & Chen, Y. (2022). A graph neural network-based node classification model on class-imbalanced graph data. Knowledge-Based Systems, 244, Article 108538. https://doi.org/10.1016/j.knosys.2022.108538
• Katz, L. (1953). A New Status Index Derived from Sociometric Analysis. Psychometrika, 18(1), 39–43. https://doi.org/10.1007/BF02289026
• Keng, Y. Y., Kwa, K. H., & McClain, C. (2021). Convex combinations of centrality measures. The Journal of Mathematical Sociology, 45(4), 195–222. https://doi.org/10.1080/0022250X.2020.1765776
• Kleinberg, J. M. (1999). Authoritative sources in a hyperlinked environment. Journal of the ACM, 46(5), 604–632. https://doi.org/10.1145/324133.324140
• Krackhardt, D. (1990). Assessing the Political Landscape: Structure, Cognition, and Power in Organizations. Administrative Science Quarterly, 35(2), 342–369. https://doi.org/10.2307/2393394
• Li, S., Quan, Y., Luo, X., & Wang, J. (2025). Influential nodes identification for complex networks based on multi-feature fusion. Scientific Reports, 15(1), Article 11440. https://doi.org/10.1038/s41598-025-94193-w
• Li, Y., Yang, Y., Zhu, J., Chen, H., & Wang, H. (2024). LLM-empowered few-shot node classification on incomplete graphs with real node degrees. In Proceedings of the 33rd ACM International Conference on Information and Knowledge Management (pp. 1306–1315). https://doi.org/10.1145/3627673.3679861
• Liu, W., Lu, P., & Zhang, T. (2024). Identifying influential nodes in complex networks from semi-local and global perspective. IEEE Transactions on Computational Social Systems, 11(2), 2105–2120. https://doi.org/10.1109/TCSS.2023.3295177
• Lyu, B., Hamdi, M., Yang, Y., Cao, Y., Yan, Z., Li, K., Wen, S., & Huang, T. (2023). Efficient Spectral Graph Convolutional Network Deployment on Memristive Crossbars. IEEE Transactions on Emerging Topics in Computational Intelligence, 7(2), 415–425. https://doi.org/10.1109/TETCI.2022.3210998
• Mohammad, B. B., Dhuli, S., & Enduri, M. K. (2025). Isolating Centrality-Based Generalization of Traditional Centralities to Discover Vital Nodes in Complex Networks. Arabian Journal for Science and Engineering, 50(15), 12003–12025. https://doi.org/10.1007/s13369-024-09628-9
• Mukhtar, M. F., Abal Abas, Z., Baharuddin, A. S., Norizan, M. N., Fakhruddin, W. F. W. W., Minato, W., Rasib, A. H. A., Abidin, Z. Z., Rahman, A. F. N. A., & Anuar, S. H. H. (2023). Integrating local and global information to identify influential nodes in complex networks. Scientific Reports, 13(1), Article 11411. https://doi.org/10.1038/s41598-023-37570-7
• Page, L., Brin, S., Motwani, R., & Winograd, T. (1999). The PageRank citation ranking: Bringing order to the web (Stanford InfoLab Technical Report No. 1999–66). Stanford University. http://ilpubs.stanford.edu:8090/422/
• Qiu, L., Zhang, J., & Tian, X. (2021). Ranking influential nodes in complex networks based on local and global structures. Applied Intelligence, 51(7), 4394–4407. https://doi.org/10.1007/s10489-020-02132-1
• Sabah, L., & Şimşek, M. (2023). A new fast entropy‐based method to generate composite centrality measures in complex networks. Concurrency and Computation: Practice and Experience, 35(10), Article e7657. https://doi.org/10.1002/cpe.7657
• Salavati, C., Abdollahpouri, A., & Manbari, Z. (2019). Ranking nodes in complex networks based on local structure and improving closeness centrality. Neurocomputing, 336, 36–45. https://doi.org/10.1016/j.neucom.2018.04.086
• Şimşek, A. (2024). Sosyal ağlarda merkezilik ölçütleri kullanılarak makine öğrenmesi ile etkili bireylerin tespiti. Yüzüncü Yıl Üniversitesi Fen Bilimleri Enstitüsü Dergisi, 29(1), 166–172. https://doi.org/10.53433/yyufbed.1348472
• Leskovec, J., & Krevl, A. (2014). SNAP datasets: Stanford large network dataset collection [Data set]. Stanford University. https://snap.stanford.edu/data/
• Ullah, A., Wang, B., Sheng, J., Long, J., Khan, N., & Sun, Z. (2021). Identification of nodes influence based on global structure model in complex networks. Scientific Reports, 11(1), Article 6173. https://doi.org/10.1038/s41598-021-84684-x
• Wang, B., Zhang, J., Dai, J., & Sheng, J. (2022). Influential nodes identification using network local structural properties. Scientific Reports, 12(1), Article 1833. https://doi.org/10.1038/s41598-022-05564-6
• Wang, S., Huang, J., Chen, Z., Song, Y., Tang, W., Mao, H., Fan, W., Liu, H., Liu, X., Yin, D., & Li, Q. (2025). Graph machine learning in the era of large language models (LLMs). ACM Transactions on Intelligent Systems and Technology, 16(5), 1–40. https://doi.org/10.1145/3732786
• Wu, Y., Hu, Y., Yin, S., Cai, B., Tang, X., & Li, X. (2024). A graph convolutional network model based on regular equivalence for identifying influential nodes in complex networks. Knowledge-Based Systems, 301, Article 112235. https://doi.org/10.1016/j.knosys.2024.112235
• Xu, X., Zhu, C., Wang, Q., Zhu, X., & Zhou, Y. (2020). Identifying vital nodes in complex networks by adjacency information entropy. Scientific Reports, 10(1), Article 2691. https://doi.org/10.1038/s41598-020-59616-w
• Yang, Y., Wang, X., Chen, Y., Hu, M., & Ruan, C. (2020). A novel centrality of influential nodes identification in complex networks. IEEE Access, 8, 58742–58751. https://doi.org/10.1109/ACCESS.2020.2983053
• Zachary, W. W. (1977). An information flow model for conflict and fission in small groups. Journal of Anthropological Research, 33(4), 452–473.
• Zhang, Y., & Xiao, X. (2022). A dynamic community detection method for complex networks based on deep self-coding network. Computational Intelligence and Neuroscience, 2022(1), Article 7084084. https://doi.org/10.1155/2022/7084084
• Zhao, G., Jia, P., Huang, C., Zhou, A., & Fang, Y. (2020). A machine learning–based framework for identifying influential nodes in complex networks. IEEE Access, 8, 65462–65471. https://doi.org/10.1109/ACCESS.2020.2984286
• Zhao, G., Jia, P., Zhou, A., & Zhang, B. (2020). InfGCN: Identifying influential nodes in complex networks with graph convolutional networks. Neurocomputing, 414, 18–26. https://doi.org/10.1016/j.neucom.2020.07.028
• Zhao, J., Wang, Y., & Deng, Y. (2020). Identifying influential nodes in complex networks from global perspective. Chaos, Solitons & Fractals, 133, Article 109637. https://doi.org/10.1016/j.chaos.2020.109637