XHBot: eXplainable Heterophily-aware Graph Neural Networks for Social Bot Detection
DOI:
https://doi.org/10.4108/airo.12969Keywords:
Social bot detection, graph neural networks, Heterophily, Explainable AI (XAI), XHBotAbstract
Social bots threaten the integrity of online ecosystems by engaging in coordinated opinion manipulation. While Graph Neural Networks (GNNs) have become a dominant paradigm for bot detection, modern camouflaged bots strategically follow benign users to evade detection, creating structural heterophily that degrades the performance of standard homophilic GNN aggregators; moreover, many existing detectors offer limited forensic explainability. To address these challenges jointly, we propose XHBot (eXplainable Heterophily-aware Bot detector), a framework that is robust to heterophilic relation camouflage while providing transparent, multi-level forensic evidence for platform moderation. XHBot couples three components: Spectral-Guided Topology Refinement (SGTR), which down-weights camouflage edges by their contribution to the graph’s high-frequency (Dirichlet) energy before aggregation; Tri-Channel Heterophily-Aware Aggregation (THCA), which separates homophilic, heterophilic, and self-identity signals; and Contrastive Prototype Disentanglement (CPD), which decouples behavioural signatures from social positioning. Evaluated on TwiBot-20, TwiBot-22, and Cresci-2017 under a unified protocol, XHBot reaches an F1 score of 0.9474 on TwiBot-20, improving over a competitive suite of recent baselines (including RGT, NeighborSense, and HW-GNN) by 9.64%. Its Hierarchical Forensic Explanation (HFE) module extracts both instance-level subgraphs and community-level diagnostic motifs, which we assess quantitatively (Fidelity, Sparsity) and through qualitative case studies. These results indicate that decoupling behavioural signatures from adversarial social positioning is valuable for modern bot detection, and that combining accuracy with interpretable evidence supports deployment in real-world moderation settings.
Downloads
References
[1] Ng LHX, Carley KM. A global comparison of social media bot and human characteristics. Sci Rep [Internet]. 2025 [cited 2026 Jul 7];15(1):10973. Available from: https://doi.org/10.1038/s41598-025-96372-1
[2] Federal Trade Commission. New FTC data show people have lost billions to social media scams [Internet].Washington (DC): Federal Trade Commission; 2026 Apr 27 [cited 2026 Jul 7]. Available from: https://www.ftc.gov/news-events/news/pressreleases/ 2026/04/new-ftc-data-show-peoplehave- lost-billions-social-media-scams
[3] Quang BHD, Trinh TT, Dang QV. Integrating graph features for social bot detection. In: Intelligence of Things: Technologies and Applications (ICIT 2025). Lecture Notes on Data Engineering and Communications Technologies, vol. 282. Springer; 2026. doi:10.1007/978- 3-032-13254-3_6
[4] Cresci S. A decade of social bot detection. Commun ACM. 2020;63(10):72-83. doi:10.1145/3409116
[5] Cresci S, Di Pietro R, Petrocchi M, Spognardi A, Tesconi M. The paradigm-shift of social spambots: evidence, theories, and tools for the arms race. In: Proceedings of the 26th International Conference on World Wide Web Companion (WWW 2017 Companion). ACM; 2017. p. 963-72. doi:10.1145/3041021.3055135
[6] Feng S, Wan H, Wang N, Li J, Luo M. TwiBot-20: a comprehensive Twitter bot detection benchmark. In: Proceedings of the 30th ACM International Conference on Information and Knowledge Management (CIKM 2021). ACM; 2021. p. 4485-94. doi:10.1145/3459637.3482019
[7] Feng S, Tan Z, Wan H, Wang N, Chen Z, Zhang B, et al. TwiBot-22: towards graph-based Twitter bot detection [Internet]. In: Advances in Neural Information Processing Systems 35 (NeurIPS 2022), Datasets and Benchmarks Track. 2022 [cited 2026 Jul 7]. Available from: https://proceedings.neurips.cc/paper_files/paper/2022/hash/e4fd610b1d77699a02df07ae97d992a-Abstract-Datasets_and_Benchmarks.html
[8] Li Y, Lu H, Chen W. Neighborhood perceivable graph neural network for relational heterogeneous Twitter bot detection. PLoS One. 2026;21. doi:10.1371/journal.pone.0342686
[9] Cheng M, Xiao Y, Huang T, Lei C, Zhang C. CBMTE: social bot detection via multi-source heterogeneous feature fusion. Sensors. 2025;25(11):3549. doi:10.3390/s25113549
[10] Liu Z, Gao J, Ji Z, Zhao L. HW-GNN: homophily-aware Gaussian-window constrained graph spectral network for social network bot detection. arXiv preprint. 2025. doi:10.48550/arXiv.2511.22493
[11] Fu C, Chen K, Pan X, Yu S, Ni J, Min Y. Social bot detection via heterogeneous graph learning and sample balancing strategies. In: Big Data and Social Computing (BDSC 2025). Communications in Computer and Information Science, vol. 2622. Springer; 2026. doi:10.1007/978-981-95-0880-8_18
[12] He B, Jiang X, Wu Q, Liu H, Yang Y, Liao Y. Boosting bot detection via heterophily-aware representation learning and prototype-guided cluster discovery. In: Proceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2025). ACM; 2025. doi:10.1145/3711896.3736862
[13] Dou Y, Liu Z, Sun L, Deng Y, Peng H, Yu PS. Enhancing graph neural network-based fraud detectors against camouflaged fraudsters. In: Proceedings of the 29th ACM International Conference on Information and Knowledge Management (CIKM 2020). ACM; 2020. doi:10.1145/3340531.3411903
[14] Zhang W, Zhong J, Yao G, Han R, Lin X, Zhang Z, et al. Dual-channel heterophilic message passing for graph fraud detection. In: 2025 International Joint Conference on Neural Networks (IJCNN); 2025. doi:10.48550/arXiv.2504.14205
[15] Li E, Ouyang J, Xiang S, Yang M, Chen Y. Efficient relation-aware heterogeneous graph neural network for fraud detection. World Wide Web. 2025;28:55. doi:10.1007/s11280-025-01369-5
[16] Schlichtkrull M, Kipf TN, Bloem P, van den Berg R, Titov I, Welling M. Modeling relational data with graph convolutional networks. In: The Semantic Web – 15th International Conference (ESWC 2018). Lecture Notes in Computer Science, vol. 10843. Springer; 2018. p. 593- 607. doi:10.1007/978-3-319-93417-4_38
[17] Feng S, Wan H, Wang N, Luo M. BotRGCN: Twitter bot detection with relational graph convolutional networks. In: Proceedings of the 2021 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM 2021). ACM; 2021. p. 236-9. doi:10.1145/3487351.3488336
[18] Dang QV, Nguyen NSA. Evaluating the contribution of relationship information in detecting fraud using graph neural networks. In: Inventive Communication and Computational Technologies: Proceedings of ICICCT 2022. Springer; 2022. p. 865-75. doi:10.1007/978-981- 19-4960-9_65
[19] Feng S, Tan Z, Li R, Luo M. Heterogeneity-aware Twitter bot detection with relational graph transformers. In: Proceedings of the 36th AAAI Conference on Artificial Intelligence (AAAI 2022). 2022;36:3977-85. doi:10.1609/aaai.v36i4.20314
[20] Qiao B, Chen Y, Li K, Zhou W, Hu S, Song Y. MGDIL: multi-granularity summarization and domain-invariant learning for cross-domain social bot detection. arXiv preprint. 2026. doi:10.48550/arXiv.2603.27928
[21] Ying R, Bourgeois D, You J, Zitnik M, Leskovec J. GNNExplainer: generating explanations for graph neural networks. In: Advances in Neural Information Processing Systems 32 (NeurIPS 2019); 2019. doi:10.48550/arXiv.1903.03894
[22] Luo D, Cheng W, Xu D, Yu W, Zong B, Chen H, et al. Parameterized explainer for graph neural network [Internet]. In: Advances in Neural Information Processing Systems 33 (NeurIPS 2020). 2020 [cited 2026 Jul 7]. Available from: https://proceedings.neurips.cc/paper/2020/hash/e37b08dd3015330dcbb5d6663667b8b8-Abstract.html
[23] Zhang L, Wang X, Du H, Xu Y, Liu Z, Liu Y. RABot: reinforcement-guided graph augmentation for imbalanced and noisy social bot detection. arXiv preprint. 2026. doi:10.48550/arXiv.2602.21749
[24] Yang Y, Liu H, Zhang X, Liu Y, Xia Y,Wu Q, et al. FedRio: personalized federated social bot detection via cooperative reinforced contrastive adversarial distillation. arXiv preprint. 2026. doi:10.48550/arXiv.2604.10678
[25] Liu S, Wang H, Li Z, Wei P. FrequencyFormer: oriented object detection with frequency transformer. EAI Endorsed Transactions on AI and Robotics. 2025;4. doi:10.4108/airo.10701
[26] Xue Z, Wang B, Xie Y, Li Z, Fan X, Lin C, et al. FDD-YOLO: a lightweight multi-scale prohibited items detection model. EAI Endorsed Transactions on AI and Robotics. 2025;4. doi:10.4108/airo.10277
[27] Musa W, Katili MR, Ridwan W. Optimizing machine learning architectures for time series forecasting: a hybrid rvGA-eNM approach. EAI Endorsed Transactions on AI and Robotics. 2025;4. doi:10.4108/airo.9976
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Quang-Vinh Dang, Phuong-Lan Nguyen, Dat Le, Minh Ngoc Dinh
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
This is an open access article distributed under the terms of the CC BY-NC-SA 4.0, which permits copying, redistributing, remixing, transformation, and building upon the material in any medium so long as the original work is properly cited.
