{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,6]],"date-time":"2026-06-06T12:04:00Z","timestamp":1780747440411,"version":"3.54.1"},"reference-count":40,"publisher":"MDPI AG","issue":"3","license":[{"start":{"date-parts":[[2025,3,10]],"date-time":"2025-03-10T00:00:00Z","timestamp":1741564800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"National Natural Science Foundation of China","award":["U19B2004"],"award-info":[{"award-number":["U19B2004"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Entropy"],"abstract":"<jats:p>There are complex graph structures and rich textual information on social networks. Text provides important information for various tasks, while graph structures offer multilevel context for the semantics of the text. Contemporary researchers tend to represent these kinds of data by text-attributed graphs (TAGs). Most TAG-based representation learning methods focus on designing frameworks that convey graph structures to large language models (LLMs) to generate semantic embeddings for downstream graph neural networks (GNNs). However, these methods only provide text attributes for nodes, which fails to capture the multilevel context and leads to the loss of valuable information. To tackle this issue, we introduce the Multilevel Context Learner (MCL) model, which leverages multilevel context on social networks to enhance LLMs\u2019 semantic embedding capabilities. We model the social network as a multilevel context textual-edge graph (MC-TEG), effectively capturing both graph structure and semantic relationships. Our MCL model leverages the reasoning capabilities of LLMs to generate semantic embeddings by integrating these multilevel contexts. The tailored bidirectional dynamic graph attention layers are introduced to further distinguish the weight information. Experimental evaluations on six real social network datasets show that the MCL model consistently outperforms all baseline models. Specifically, the MCL model achieves prediction accuracies of 77.98%, 77.63%, 74.61%, 76.40%, 72.89%, and 73.40%, with absolute improvements of 9.04%, 9.19%, 11.05%, 7.24%, 6.11%, and 9.87% over the next best models. These results demonstrate the effectiveness of the proposed MCL model.<\/jats:p>","DOI":"10.3390\/e27030286","type":"journal-article","created":{"date-parts":[[2025,3,10]],"date-time":"2025-03-10T06:59:59Z","timestamp":1741589999000},"page":"286","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":5,"title":["Multilevel Context Learning with Large Language Models for Text-Attributed Graphs on Social Networks"],"prefix":"10.3390","volume":"27","author":[{"given":"Xiaokang","family":"Cai","sequence":"first","affiliation":[{"name":"Electronic Information School, Wuhan University, Wuhan 430072, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ruoyuan","family":"Gong","sequence":"additional","affiliation":[{"name":"Electronic Information School, Wuhan University, Wuhan 430072, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hao","family":"Jiang","sequence":"additional","affiliation":[{"name":"Electronic Information School, Wuhan University, Wuhan 430072, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2025,3,10]]},"reference":[{"key":"ref_1","unstructured":"Gottfried, J., and Shearer, E. (2024, June 03). News Use Across Social Media Platforms 2016. Pew Research Center, 2016. Available online: https:\/\/apo.org.au\/node\/64483."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"122271","DOI":"10.1016\/j.techfore.2022.122271","article-title":"Research on multi-label user classification of social media based on ML-KNN algorithm","volume":"188","author":"Huang","year":"2023","journal-title":"Technol. Forecast. Soc. Chang."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"e12634","DOI":"10.1111\/exsy.12634","article-title":"Recommendation of users in social networks: A semantic and social based classification approach","volume":"38","author":"Berkani","year":"2021","journal-title":"Expert Syst."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"4682","DOI":"10.1109\/TNNLS.2021.3137396","article-title":"A comprehensive survey on community detection with deep learning","volume":"35","author":"Su","year":"2024","journal-title":"IEEE Trans. Neural Netw. Learn. Syst."},{"key":"ref_5","unstructured":"Zeng, H., Zhou, H., Srivastava, A., Kannan, R., and Prasanna, V. (2019). Graphsaint: Graph sampling based inductive learning method. arXiv."},{"key":"ref_6","unstructured":"Jiang, J., and Ferrara, E. (2023). Social-LLM: Modeling User Behavior at Scale using Language Models and Social Network Data. arXiv."},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Xie, B., Ma, X., Shan, X., Beheshti, A., Yang, J., Fan, H., and Wu, J. (2024). Multiknowledge and LLM-Inspired Heterogeneous Graph Neural Network for Fake News Detection. IEEE Trans. Comput. Soc. Syst.","DOI":"10.1109\/TCSS.2024.3488191"},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Bai, H., Chen, Z., Lyu, M.R., King, I., and Xu, Z. (2018, January 22\u201326). Neural relational topic models for scientific article analysis. Proceedings of the 27th ACM International Conference on Information and Knowledge Management, Torino, Italy.","DOI":"10.1145\/3269206.3271696"},{"key":"ref_9","first-page":"1","article-title":"Graph neural collaborative topic model for citation recommendation","volume":"40","author":"Xie","year":"2021","journal-title":"ACM Trans. Inf. Syst. (TOIS)"},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Ma, Z., Dou, Z., Xu, W., Zhang, X., Jiang, H., Cao, Z., and Wen, J.R. (2021, January 1\u20135). Pre-training for ad-hoc retrieval: Hyperlink is also you need. Proceedings of the 30th ACM International Conference on Information & Knowledge Management, Virtual Event.","DOI":"10.1145\/3459637.3482286"},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"43","DOI":"10.1007\/s13042-010-0001-0","article-title":"Understanding bag-of-words model: A statistical framework","volume":"1","author":"Zhang","year":"2010","journal-title":"Int. J. Mach. Learn. Cybern."},{"key":"ref_12","unstructured":"Mikolov, T., Sutskever, I., Chen, K., Corrado, G.S., and Dean, J. (2013, January 5\u201310). Distributed representations of words and phrases and their compositionality. Proceedings of the Advances in Neural Information Processing Systems 26 (NIPS 2013), Lake Tahoe, NV, USA."},{"key":"ref_13","unstructured":"Duan, K., Liu, Q., Chua, T.S., Yan, S., Ooi, W.T., Xie, Q., and He, J. (2023). Simteg: A frustratingly simple approach improves textual graph learning. arXiv."},{"key":"ref_14","unstructured":"He, X., Bresson, X., Laurent, T., Perold, A., LeCun, Y., and Hooi, B. (2023). Harnessing explanations: Llm-to-lm interpreter for enhanced text-attributed graph representation learning. arXiv."},{"key":"ref_15","unstructured":"Qin, Y., Wang, X., Zhang, Z., and Zhu, W. (2023). Disentangled representation learning with large language models for text-attributed graphs. arXiv."},{"key":"ref_16","unstructured":"Brody, S., Alon, U., and Yahav, E. (2021). How attentive are graph attention networks?. arXiv."},{"key":"ref_17","unstructured":"Yang, J., Liu, Z., Xiao, S., Li, C., Lian, D., Agrawal, S., Singh, A., Sun, G., and Xie, X. (2021, January 6\u201314). Graphformers: Gnn-nested transformers for representation learning on textual graph. Proceedings of the Advances in Neural Information Processing Systems 34 (NeurIPS 2021), Virtual."},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Wang, X., Cui, P., Wang, J., Pei, J., Zhu, W., and Yang, S. (2017, January 4\u20139). Community preserving network embedding. Proceedings of the AAAI Conference on Artificial Intelligence, San Francisco, CA, USA.","DOI":"10.1609\/aaai.v31i1.10488"},{"key":"ref_19","doi-asserted-by":"crossref","unstructured":"Wang, X., Ji, H., Shi, C., Wang, B., Ye, Y., Cui, P., and Yu, P.S. (2019, January 13\u201317). Heterogeneous graph attention network. Proceedings of the World Wide Web Conference, San Francisco, CA, USA.","DOI":"10.1145\/3308558.3313562"},{"key":"ref_20","unstructured":"Kipf, T.N., and Welling, M. (2016). Semi-supervised classification with graph convolutional networks. arXiv."},{"key":"ref_21","unstructured":"Hamilton, W., Ying, Z., and Leskovec, J. (2017, January 4\u20139). Inductive representation learning on large graphs. Proceedings of the Advances in Neural Information Processing Systems 30 (NIPS 2017), Long Beach, CA, USA."},{"key":"ref_22","unstructured":"Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A.N., Kaiser, \u0141., and Polosukhin, I. (2017, January 4\u20139). Attention is all you need. Proceedings of the Advances in Neural Information Processing Systems 30 (NIPS 2017), Long Beach, CA, USA."},{"key":"ref_23","unstructured":"Srivastava, A., and Sutton, C. (2017). Autoencoding variational inference for topic models. arXiv."},{"key":"ref_24","doi-asserted-by":"crossref","unstructured":"Zhang, D.C., and Lauw, H.W. (2021, January 1\u20135). Topic modeling for multi-aspect listwise comparisons. Proceedings of the 30th ACM International Conference on Information & Knowledge Management, Virtual Event.","DOI":"10.1145\/3459637.3482398"},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Zhu, J., Cui, Y., Liu, Y., Sun, H., Li, X., Pelger, M., Yang, T., Zhang, L., Zhang, R., and Zhao, H. (2021, January 19\u201323). Textgnn: Improving text encoder via graph neural network in sponsored search. Proceedings of the Web Conference 2021, Ljubljana, Slovenia.","DOI":"10.1145\/3442381.3449842"},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Li, C., Pang, B., Liu, Y., Sun, H., Liu, Z., Xie, X., Yang, T., Cui, Y., Zhang, L., and Zhang, Q. (2021, January 11\u201315). Adsgnn: Behavior-graph augmented relevance modeling in sponsored search. Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval, Virtual Event.","DOI":"10.1145\/3404835.3462926"},{"key":"ref_27","unstructured":"Zhang, C., and Lauw, H.W. (2020, January 7\u201312). Topic modeling on document networks with adjacent-encoder. Proceedings of the AAAI Conference on Artificial Intelligence, New York, NY, USA."},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"1328","DOI":"10.1109\/TKDE.2023.3303465","article-title":"Topic Modeling on Document Networks with Dirichlet Optimal Transport Barycenter","volume":"36","author":"Zhang","year":"2023","journal-title":"IEEE Trans. Knowl. Data Eng."},{"key":"ref_29","unstructured":"Wang, H., Feng, S., He, T., Tan, Z., Han, X., and Tsvetkov, Y. (2023, January 10\u201316). Can language models solve graph problems in natural language?. Proceedings of the Advances in Neural Information Processing Systems 36 (NIPS 2023), New Orleans, LA, USA."},{"key":"ref_30","unstructured":"Guo, J., Du, L., Liu, H., Zhou, M., He, X., and Han, S. (2023). Gpt4graph: Can large language models understand graph structured data? an empirical evaluation and benchmarking. arXiv."},{"key":"ref_31","unstructured":"Ye, R., Zhang, C., Wang, R., Xu, S., and Zhang, Y. (2023). Language is All a Graph Needs. arXiv."},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"61","DOI":"10.1109\/TNN.2008.2005605","article-title":"The graph neural network model","volume":"20","author":"Scarselli","year":"2008","journal-title":"IEEE Trans. Neural Netw."},{"key":"ref_33","unstructured":"Mnih, V., Heess, N., and Graves, A. (2014, January 8\u201313). Recurrent models of visual attention. Proceedings of the Advances in Neural Information Processing Systems 27 (NIPS 2014), Montreal, QC, Canada."},{"key":"ref_34","unstructured":"Veli\u010dkovi\u0107, P., Cucurull, G., Casanova, A., Romero, A., Lio, P., and Bengio, Y. (2017). Graph attention networks. arXiv."},{"key":"ref_35","doi-asserted-by":"crossref","unstructured":"Xue, L., Constant, N., Roberts, A., Kale, M., Al-Rfou, R., Siddhant, A., Barua, A., and Raffel, C. (2020). mT5: A massively multilingual pre-trained text-to-text transformer. arXiv.","DOI":"10.18653\/v1\/2021.naacl-main.41"},{"key":"ref_36","unstructured":"He, P., Liu, X., Gao, J., and Chen, W. (2020). Deberta: Decoding-enhanced bert with disentangled attention. arXiv."},{"key":"ref_37","unstructured":"Touvron, H., Lavril, T., Izacard, G., Martinet, X., Lachaux, M.A., Lacroix, T., Rozi\u00e8re, B., Goyal, N., Hambro, E., and Azhar, F. (2023). Llama: Open and efficient foundation language models. arXiv."},{"key":"ref_38","unstructured":"Glm, T., Zeng, A., Xu, B., Wang, B., Zhang, C., Yin, D., Zhang, D., Rojas, D., Feng, G., and Zhao, H. (2024). Chatglm: A family of large language models from glm-130b to glm-4 all tools. arXiv."},{"key":"ref_39","unstructured":"Bai, J., Bai, S., Chu, Y., Cui, Z., Dang, K., Deng, X., Fan, Y., Ge, W., Han, Y., and Huang, F. (2023). Qwen Technical Report. arXiv."},{"key":"ref_40","unstructured":"Sun, Y., Wang, S., Feng, S., Ding, S., Pang, C., Shang, J., Liu, J., Chen, X., Zhao, Y., and Lu, Y. (2021). Ernie 3.0: Large-scale knowledge enhanced pre-training for language understanding and generation. arXiv."}],"container-title":["Entropy"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1099-4300\/27\/3\/286\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,9]],"date-time":"2025-10-09T16:49:48Z","timestamp":1760028588000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1099-4300\/27\/3\/286"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,3,10]]},"references-count":40,"journal-issue":{"issue":"3","published-online":{"date-parts":[[2025,3]]}},"alternative-id":["e27030286"],"URL":"https:\/\/doi.org\/10.3390\/e27030286","relation":{},"ISSN":["1099-4300"],"issn-type":[{"value":"1099-4300","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,3,10]]}}}