An Inference Framework of Markov Logic Network for Link Prediction in Heterogeneous Networks
Source abstract
Abstract
The presence of multiplex edges and sparse links often hampers the efficacy of link prediction (LP) tasks. By harnessing the expressive power of Markov logic network (MLN) formulations, multiplex edges can be unified to enhance LP effectiveness. However, scaling up inferences for effective LP remains challenging due to the inefficiency of traditional MLN inference methods. To tackle this issue, we redefine LP tasks within heterogeneous networks using MLN inferences and introduce a tailored inference framework to handle unobserved nodes and complex MLN structures. We propose a method to partition the MLN structure into discrete substructures and compute node label distributions using the variational expectation maximization (VEM) algorithm. Additionally, we establish a termination condition to streamline inference search space and present the MLN-based LP algorithm. Experimental findings demonstrate the efficacy of our VEM-driven MLN inference framework for LP tasks in heterogeneous networks, showcasing superior accuracy compared to existing approaches.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Advanced Graph Neural Networks · Bayesian Modeling and Causal Inference · Complex Network Analysis Techniques
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
School of Information Science and Engineering, Yunnan University, Kunming 650500, China
Yunnan Key Laboratory of Intelligent Systems and Computing, Yunnan University, Kunming 650500, China
Funding
Key Program of Basic Research of Yunnan Province · Program of Yunnan Key Laboratory of Intelligent Systems and Computing · Open Foundation of Yunnan Key Laboratory of Intelligent Control and Application
References (33)
Show source-provided references
1. Link prediction by continuous spatiotemporal representation via neural differential equations10.1016/j.knosys.2024.111619 ↗
2. A multi-type transferable method for missing link prediction in heterogeneous social networks10.1109/TKDE.2022.3233481 ↗
3. Knowledge graph embedding in a uniform space10.3233/IDA-227123 ↗
4. Community preserving adaptive graph convolutional networks for link prediction in attributed networks10.1016/j.knosys.2023.110589 ↗
5. Asymmetric multilevel interactive attention network integrating reviews for item recommendation10.3233/IDA-230128 ↗
6. TDAN: Transferable domain adversarial network for link prediction in heterogeneous social networks
7. Nguyen, T., Liu, Z., and Fang, Y. (May, January 30). Link prediction on latent heterogeneous graphs. Proceedings of the ACM Web Conference 2023, Austin, TX, USA.10.1145/3543507.3583284 ↗
8. Zhang, S., Zhang, J., Song, X., and Adeshina, S. (May, January 30). PaGE-Link: Path-based graph neural network explanation for heterogeneous link prediction. Proceedings of the ACM Web Conference 2023, Austin, TX, USA.10.1145/3543507.3583511 ↗
9. A hybrid driving decision-making system integrating markov logic networks and connectionist AI10.1109/TITS.2022.3227122 ↗
10. Unifying logical and statistical AI with Markov logic10.1145/3241978 ↗
11. Qu, M., and Tang, J. (2019, January 8–14). Probabilistic logic neural networks for reasoning. Proceedings of the 33rd International Conference on Neural Information Processing Systems, Vancouver, BC, Canada.
12. Zhang, M., and Chen, Y. (2018, January 3–8). Link prediction based on graph neural networks. Proceedings of the 32nd International Conference on Neural Information Processing Systems, Montreal, QC, Canada.
13. Song, K., Yue, K., Duan, L., Yang, M., and Li, A. (2022, January 1–5). Mutual information based Bayesian graph neural network for few-shot learning. Proceedings of the 38th Conference on Uncertainty in Artificial Intelligence, PMLR, Eindhoven, The Netherlands.
14. Link Prediction Method Combining Node Labels with Common Neighbors
15. Wang, X., Yang, H., and Zhang, M. (2023). Neural common neighbor with completion for link prediction. arXiv.
16. Neural graph embeddings as explicit low-rank matrix factorization for link prediction10.1016/j.patcog.2022.108977 ↗
17. MHGCN+: Multiplex Heterogeneous Graph Convolutional Network
18. Zhang, J., Wei, L., Xu, Z., and Yao, Q. (2024, January 25–29). Heuristic Learning with Graph Neural Networks: A Unified Framework for Link Prediction. Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, Barcelona, Spain.10.1145/3637528.3671946 ↗
19. Dimitriou, P., and Karyotis, V. (2023). A combinatory framework for link prediction in complex networks. Appl. Sci., 13.10.3390/app13179685 ↗
20. Feng, P., Zhang, X., Wu, H., Wang, Y., Yang, Z., and Ouyang, D. (2024). Link Prediction Based on Feature Mapping and Bi-Directional Convolution. Appl. Sci., 14.10.3390/app14052089 ↗
21. Jin, B., Zhang, Y., Zhu, Q., and Han, J. (2023, January 6–10). Heterformer: Transformer-based deep node representation learning on heterogeneous text-rich networks. Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, Long Beach, CA, USA.10.1145/3580305.3599376 ↗
22. Zhan, Q., Wang, J., Li, Y., Xie, Z., and Liu, Y. (2024, January 5–8). HRMNN: Heterogeneous Relationship Mined Graph Neural Network. Proceedings of the 12th International Conference on Intelligent Computing, Tianjin, China.10.1007/978-981-97-5618-6_4 ↗
23. Chen, F., Weitkämper, F., and Malhotra, S. (2024, January 8–12). Understanding Domain-Size Generalization in Markov Logic Networks. Proceedings of the 34th Joint European Conference on Machine Learning and Knowledge Discovery in Databases, Vilnius, Lithuania.10.1007/978-3-031-70368-3_18 ↗
24. Wasserman, M., and Mateos, G. (2024). Graph structure learning with interpretable Bayesian neural networks. arXiv.
25. Quantified neural Markov logic networks10.1016/j.ijar.2024.109172 ↗
26. Harsha, V.L., Jia, G., and Kok, S. (2023, January 20–24). Probabilistic logic graph attention networks for reasoning. Proceedings of the 29th ACM Web Conference, ACM, Taiwan, China.
27. Fang, H., Liu, Y., Cai, Y., and Sun, M. (May, January 30). MLN4KB: An efficient markov logic network engine for large-scale knowledge bases and structured logic rules. Proceedings of the ACM Web Conference 2023, Austin, TX, USA.10.1145/3543507.3583248 ↗
28. MCLA: Research on cumulative learning of Markov Logic Network10.1016/j.knosys.2022.108352 ↗
29. Qu, M., Chen, J., Xhonneux, L., Bengio, Y., and Tang, J. (2021, January 3–7). RNNLogic: Learning Logic Rules for Reasoning on Knowledge Graphs. Proceedings of the 9th International Conference on Learning Representations, Virtual.
30. Zhang, Y., Chen, X., Yang, Y., Ramamurthy, A., Li, B., Qi, Y., and Song, L. (2020, January 26–30). Efficient Probabilistic Logic Reasoning with Graph Neural Networks. Proceedings of the 8th International Conference on Learning Representations(ICLR), Addis Ababa, Ethiopia.
31. Markov logic networks10.1007/s10994-006-5833-1 ↗
32. Qu, M., Bengio, Y., and Tang, J. (2019, January 9–15). Gmnn: Graph markov neural networks. Proceedings of the 15th International Conference on Machine Learning, PMLR, Long Beach, CA, USA.
33. Marra, G., and Kuželka, O. (2021, January 27–30). Neural markov logic networks. Proceedings of the 37th Uncertainty in Artificial Intelligence, PMLR, Online.