摘要学术文献特征表示，是学术文献搜索、分类组织、个性化推荐等学术大数据服务的关键步骤。研究表明，图神经网络能够有效学习文献的特征表示，然而当前研究主要集中在有监督学习方法上，不仅对数据集的大小和质量的要求较高，且学习到的文献特征表示与具体任务高度耦合。基于此，本文将四种无监督图神经网络方法引入学术文献表示学习，从Cora、CiteSeer和DBLP（database systems and logic programming）数据集的引文网络、共被引网络和文献耦合网络中学习文献的表示向量，并应用于文献分类和论文推荐两大下游任务。研究结果表明，①深度互信息图神经网络适合于文献分类任务，对抗正则化变分图自编码器则在论文推荐任务上性能更佳；②Cora数据集上的结果表明，相较于共被引和文献耦合网络，引文网络更适合于学习通用的文献表示向量。
1 Landhuis E. Scientific literature: information overload[J]. Nature, 2016, 535(7612): 457-458. 2 Sulova S, Todoranova L, Penchev B, et al. Using text mining to classify research papers[C]// Proceedings of the 17th International Multidisciplinary Scientific GeoConference, Bulgaria, Albena, 2017: 647-654. 3 Chandrasekaran K, Gauch S, Lakkaraju P, et al. Concept-based document recommendations for CiteSeer authors[C]// Proceedings of the International Conference on Adaptive Hypermedia and Adaptive Web-Based Systems. Heidelberg: Springer, 2008: 83-92. 4 Sun C, Qiu X P, Xu Y G, et al. How to fine-tune BERT for text classification?[C]// Proceedings of the China National Conference on Chinese Computational Linguistics. Cham: Springer, 2019: 194-206. 5 Hassan H, Sansonetti G, Gasparetti F, et al. BERT, ELMo, USE and InferSent sentence encoders: the panacea for research-paper recommendation?[C]// Proceedings of the 13th ACM Conference on Recommender Systems. New York: ACM Press, 2019: 6-10. 6 Kong X J, Mao M Y, Wang W, et al. VOPRec: vector representation learning of papers with text information and structural identity for recommendation[J]. IEEE Transactions on Emerging Topics in Computing, 2021, 9(1): 226-237. 7 刘欢, 李晓戈, 胡立坤, 等. 基于知识图谱驱动的图神经网络推荐模型[J]. 计算机应用, 2021, 41(7): 1865-1870. 8 吴国栋, 查志康, 涂立静, 等. 图神经网络推荐研究进展[J]. 智能系统学报, 2020, 15(1): 14-24. 9 Wu Z H, Pan S R, Chen F W, et al. A comprehensive survey on graph neural networks[J]. IEEE Transactions on Neural Networks and Learning Systems, 2021, 32(1): 4-24. 10 Reid Turner C, Fuggetta A, Lavazza L, et al. A conceptual basis for feature engineering[J]. Journal of Systems and Software, 1999, 49(1): 3-15. 11 Bengio Y, Courville A, Vincent P. Representation learning: a review and new perspectives[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013, 35(8): 1798-1828. 12 Krizhevsky A, Sutskever I, Hinton G. ImageNet classification with deep convolutional neural networks[C]// Proceedings of the 25th International Conference on Neural Information Processing Systems. New York: ACM Press, 2012: 1097-1105. 13 Noda K, Yamaguchi Y, Nakadai K, et al. Audio-visual speech recognition using deep learning[J]. Applied Intelligence, 2015, 42(4): 722-737. 14 Church K W. Word2Vec[J]. Natural Language Engineering, 2017, 23(1): 155-162. 15 Peters M E, Neumann M, Iyyer M, et al. Deep contextualized word representations[C]// Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. Stroudsburg: Association for Computational Linguistics, 2018: 2227-2237. 16 Devlin J, Chang M W, Lee K, et al. BERT: pre-training of deep bidirectional transformers for language understanding[C]// Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. Stroudsburg: Association for Computational Linguistics, 2019: 4171-4186. 17 Beltagy I, Lo K, Cohan A. SciBERT: a pretrained language model for scientific text[C]// Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing. Stroudsburg: Association for Computational Linguistics, 2019: 3615-3620. 18 王健宗, 孔令炜, 黄章成, 等. 图神经网络综述[J]. 计算机工程, 2021, 47(4): 1-12. 19 文贵华, 江丽君, 文军. 邻域参数动态变化的局部线性嵌入[J]. 软件学报, 2008, 19(7): 1666-1673. 20 戴志波, 王靖. 鲁棒拉普拉斯特征映射算法[J]. 计算机应用研究, 2011, 28(9): 3249-3252. 21 Ahmed A, Shervashidze N, Narayanamurthy S, et al. Distributed large-scale natural graph factorization[C]// Proceedings of the 22nd International Conference on World Wide Web. New York: ACM Press, 2013: 37-48. 22 Perozzi B, Al-Rfou R, Skiena S. DeepWalk: online learning of social representations[C]// Proceedings of the 20th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York: ACM Press, 2014: 701-710. 23 Grover A, Leskovec J. node2vec: scalable feature learning for networks[C]// Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York: ACM Press, 2016: 855-864. 24 马扬, 程光权, 梁星星, 等. 有向加权网络中的改进SDNE算法[J]. 计算机科学, 2020, 47(4): 233-237. 25 Kipf T N, Welling M. Semi-supervised classification with graph convolutional networks[C]// Proceedings of the International Conference on Learning Representations, Toulon, France, April 24 - 26, 2017. 26 Kipf T N, Welling M. Variational graph auto-encoders[OL]. (2016-11-21) [2021-03-16]. https://arxiv.org/pdf/1611.07308v1.pdf. 27 Pan S R, Hu R Q, Long G D, et al. Adversarially regularized graph autoencoder for graph embedding[C]// Proceedings of the 27th International Joint Conference on Artificial Intelligence. Palo Alto: AAAI Press, 2018: 2609-2615. 28 Veli?kovi? P, Fedus W, Hamilton W L, et al. Deep graph infomax[OL]. (2018-12-21) [2021-03-16]. https://arxiv.org/pdf/1809.10341v2.pdf. 29 Le Q, Mikolov T. Distributed representations of sentences and documents[C]// Proceedings of the 31st International Conference on International Conference on Machine Learning. JMLR.org, 2014: II-1188-II-1196. 30 Ganguly S, Pudi V. Paper2vec: combining graph and text information for scientific paper representation[C]// Proceedings of the European Conference on Information Retrieval. Cham: Springer, 2017: 383-395.
