在醫學領域中，知識圖譜在提高診斷的準確度、疾病預測及藥物發現等任務中扮演著越來越重要的角色。知識圖譜推理(Knowledge Graph Reasoning)能夠利用已知的關係去推測未知的關係，但在先前研究上，多數是直接使用整個龐大的知識圖譜實作，此方式在進行需要歸納能力的任務時，可能較無法處理未知的實體或關係。因此，本研究提出了利用分群後的子圖進行知識圖譜推理的方法，以改善鏈結預測任務的準確度。
本研究使用了多種方法對知識圖譜進行嵌入表示，並結合不同的分群方法進行子圖的萃取，通過在實驗中比較不同的嵌入技術及分群方法，證實了本研究所提之方法在鏈結預測任務上的有效性。
研究結果顯示，我們的方法不僅提高了鏈結預測的準確性，同時也展現了在大規模知識圖譜中進行有效分群的能力。未來的研究上，將繼續探索更多的嵌入技術及分群方法，以進一步提高知識圖譜推理的性能及應用範圍。

In the medical field, knowledge graphs play an increasingly important role in tasks such as improving diagnostic accuracy, predicting diseases, and discovering drugs. Knowledge Graph Reasoning can use known relationships to infer unknown ones, but previous studies have mostly used the entire vast knowledge graph directly. This approach may not handle unknown entities or relationships well when performing tasks that require inductive reasoning. Therefore, this study proposes a method of reasoning with knowledge graphs using subgraphs classified post-categorization to improve the accuracy of link prediction tasks.
This study utilized various methods to embed representations of the knowledge graph and combined different clustering methods to extract subgraphs. By comparing different embedding techniques and clustering methods in experiments, this study confirmed the effectiveness of the proposed methods in link prediction tasks.
The results show that our method not only improves the accuracy of link prediction but also demonstrates the ability to effectively categorize within large-scale knowledge graphs. Future research will continue to explore more embedding techniques and clustering methods to further enhance the performance and application scope of knowledge graph reasoning.

論文審定書 i
摘要 ii
Abstract iii
目錄 iv
圖目錄 vii
表目錄 viii
1 第一章 緒論 1
1.1 研究背景 1
1.2 研究動機 2
1.3 研究目的 2
2 第二章 文獻探討 3
2.1 知識圖譜嵌入方法 3
2.1.1 GNN(Graph Neural Network) 3
2.1.2 DeepWalk 4
2.1.3 Transformer 5
2.1.4 RotatE 7
2.1.5 TransE 7
2.2 知識圖譜之子圖應用 7
2.2.1 GraIL 7
2.2.2 GraphSAGE 8
2.2.3 SEAL 9
2.3 分群方法 9
2.3.1 K-means 9
2.3.2 BIRCH(Balanced Iterative Reducing and Clustering using Hierarchies) 10
2.3.3 LDA 11
3 第三章 研究方法 13
3.1 知識圖譜特徵表示方法 13
3.1.1 文字特徵表示 14
3.1.2 Embedding特徵表示 15
3.2 子圖萃取方法 17
3.2.1 Entity Clustering 17
3.2.2 Topic Modeling 19
3.3 子圖驗證任務 19
3.3.1 鏈結預測 19
3.4 子圖評估方法 20
3.4.1 HIT@N 21
3.4.2 MRR(Mean Reciprocal Rank) 21
4 第四章 實驗結果 23
4.1 實驗流程與設計 23
4.2 資料集介紹 23
4.2.1 ChEBI 24
4.2.2 Go 24
4.3 實驗參數設置 24
4.4 K-means分群實驗結果 25
4.4.1 基於GNN之圖嵌入表示 25
4.4.2 基於DeepWalk 之圖嵌入表示 26
4.4.3 基於Transformer 之嵌入表示 27
4.4.4 基於RotatE之嵌入表示 30
4.4.5 基於TransE之嵌入表示 31
4.4.6 小結 32
4.5 其他分群方法驗證 33
4.5.1 BIRCH分群 33
4.5.2 隨機分群 34
4.5.3 基於LDA分群 35
4.5.4 小結 37
4.6 Go資料集實驗結果 37
4.7 消融性實驗 38
4.7.1 K-means群數選擇 38
4.7.2 參數探討 39
4.8 實驗結果探討 41
5 第五章 結論 43
5.1 結論 43
5.2 未來展望 43
參考文獻 45

[1] K. Teru, E. Denis, and W. Hamilton, "Inductive relation prediction by subgraph reasoning," in International Conference on Machine Learning, 2020: PMLR, pp. 9448-9457.
[2] S. Arora, "A survey on graph neural networks for knowledge graph completion," arXiv preprint arXiv:2007.12374, 2020.
[3] R. Li et al., "How does knowledge graph embedding extrapolate to unseen data: a semantic evidence view," in Proceedings of the AAAI conference on artificial intelligence, 2022, vol. 36, no. 5, pp. 5781-5791.
[4] X. Lin, Z. Quan, Z.-J. Wang, T. Ma, and X. Zeng, "KGNN: Knowledge Graph Neural Network for Drug-Drug Interaction Prediction," in IJCAI, 2020, vol. 380, pp. 2739-2745.
[5] D. Tena Cucala, B. Cuenca Grau, E. V. Kostylev, and B. Motik, "Explainable GNN-based models over knowledge graphs," 2022.
[6] M. Zhang and Y. Chen, "Link prediction based on graph neural networks," Advances in neural information processing systems, vol. 31, 2018.
[7] B. Perozzi, R. Al-Rfou, and S. Skiena, "Deepwalk: Online learning of social representations," in Proceedings of the 20th ACM SIGKDD international conference on Knowledge discovery and data mining, 2014, pp. 701-710.
[8] A. Grover and J. Leskovec, "node2vec: Scalable feature learning for networks," in Proceedings of the 22nd ACM SIGKDD international conference on Knowledge discovery and data mining, 2016, pp. 855-864.
[9] A. Vaswani et al., "Attention is all you need," Advances in neural information processing systems, vol. 30, 2017.
[10] J. Devlin, M.-W. Chang, K. Lee, and K. Toutanova, "Bert: Pre-training of deep bidirectional transformers for language understanding," arXiv preprint arXiv:1810.04805, 2018.
[11] L. Yao, C. Mao, and Y. Luo, "KG-BERT: BERT for knowledge graph completion," arXiv preprint arXiv:1909.03193, 2019.
[12] C. Ying et al., "Do transformers really perform badly for graph representation?," Advances in neural information processing systems, vol. 34, pp. 28877-28888, 2021.
[13] Z. Sun, Z.-H. Deng, J.-Y. Nie, and J. Tang, "Rotate: Knowledge graph embedding by relational rotation in complex space," arXiv preprint arXiv:1902.10197, 2019.
[14] J. Wang, S. Zhang, R. Li, G. Chen, S. Yan, and L. Ma, "Multi-view feature representation and fusion for drug-drug interactions prediction," BMC bioinformatics, vol. 24, no. 1, p. 93, 2023.
[15] A. Bordes, N. Usunier, A. Garcia-Duran, J. Weston, and O. Yakhnenko, "Translating embeddings for modeling multi-relational data," Advances in neural information processing systems, vol. 26, 2013.
[16] W. Hamilton, Z. Ying, and J. Leskovec, "Inductive representation learning on large graphs," Advances in neural information processing systems, vol. 30, 2017.
[17] M. Ahmed, R. Seraj, and S. M. S. Islam, "The k-means algorithm: A comprehensive survey and performance evaluation," Electronics, vol. 9, no. 8, p. 1295, 2020.
[18] T. Zhang, R. Ramakrishnan, and M. Livny, "BIRCH: an efficient data clustering method for very large databases," ACM sigmod record, vol. 25, no. 2, pp. 103-114, 1996.
[19] H. Jelodar et al., "Latent Dirichlet allocation (LDA) and topic modeling: models, applications, a survey," Multimedia Tools and Applications, vol. 78, pp. 15169-15211, 2019.
[20] C. a. A. Min, Jinhyun and Lee, Taewhi and Im, Dong-Hyuk, "TK-BERT: Effective Model of Language Representation using Topic-based Knowledge Graphs," pp. 1-4, 2023, doi: 10.1109/IMCOM56909.2023.10035573.
[21] K. Ethayarajh, "How contextual are contextualized word representations? Comparing the geometry of BERT, ELMo, and GPT-2 embeddings," arXiv preprint arXiv:1909.00512, 2019.
[22] T. N. Kipf and M. Welling, "Semi-supervised classification with graph convolutional networks," arXiv preprint arXiv:1609.02907, 2016.
[23] J. Zhou et al., "Graph neural networks: A review of methods and applications," AI open, vol. 1, pp. 57-81, 2020.
[24] W. L. Hamilton, Graph representation learning. Morgan & Claypool Publishers, 2020.
[25] A. Rossi, D. Barbosa, D. Firmani, A. Matinata, and P. Merialdo, "Knowledge graph embedding for link prediction: A comparative analysis," ACM Transactions on Knowledge Discovery from Data (TKDD), vol. 15, no. 2, pp. 1-49, 2021.
[26] K. Degtyarenko et al., "ChEBI: a database and ontology for chemical entities of biological interest," Nucleic acids research, vol. 36, no. suppl_1, pp. D344-D350, 2007.
[27] M. Ashburner et al., "Gene ontology: tool for the unification of biology," Nature genetics, vol. 25, no. 1, pp. 25-29, 2000.
[28] M. Ali et al., "PyKEEN 1.0: a python library for training and evaluating knowledge graph embeddings," Journal of Machine Learning Research, vol. 22, no. 82, pp. 1-6, 2021.