CANE: Context-Aware Network Embedding for Relation Modeling

Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics, pages 1722–1731 Vancouver, Canada, July 30 - August 4, 2017. c⃝2017 Association for Computational Linguistics https://doi.org/10.18653/v1/P17-1158 Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics, pages 1722–1731 Vancouver, Canada, July 30 - August 4, 2017. c⃝2017 Association for Computational Linguistics https://doi.org/10.18653/v1/P17-1158 CANE: Context-Aware Network Embedding for Relation Modeling Cunchao Tu1,2∗, Han Liu3∗, Zhiyuan Liu1,2†, Maosong Sun1,2 1Department of Computer Science and Technology, State Key Lab on Intelligent Technology and Systems, National Lab for Information Science and Technology, Tsinghua University, China 2Jiangsu Collaborative Innovation Center for Language Ability, Jiangsu Normal University, China 3Northeastern University, China Abstract Network embedding (NE) is playing a critical role in network analysis, due to its ability to represent vertices with ef- ficient low-dimensional embedding vec- tors. However, existing NE models aim to learn a fixed context-free embedding for each vertex and neglect the diverse roles when interacting with other vertices. In this paper, we assume that one vertex usu- ally shows different aspects when interact- ing with different neighbor vertices, and should own different embeddings respec- tively. Therefore, we present Context- Aware Network Embedding (CANE), a novel NE model to address this issue. CANE learns context-aware embeddings for vertices with mutual attention mecha- nism and is expected to model the seman- tic relationships between vertices more precisely. In experiments, we compare our model with existing NE models on three real-world datasets. Experimen- tal results show that CANE achieves sig- nificant improvement than state-of-the-art methods on link prediction and compara- ble performance on vertex classification. The source code and datasets can be ob- tained from https://github.com/ thunlp/CANE. 1 Introduction Network embedding (NE), i.e., network represen- tation learning (NRL), aims to map vertices of a network into a low-dimensional space according to their structural roles in the network. NE pro- vides an efficient and effective way to represent ∗Indicates equal contribution †Corresponding Author: Z. Liu (liuzy@tsinghua.edu.cn) and manage large-scale networks, alleviating the computation and sparsity issues of conventional symbol-based representations. Hence, NE is at- tracting many research interests in recent years (Perozzi et al., 2014; Tang et al., 2015; Grover and Leskovec, 2016), and achieves promising perfor- mance on many network analysis tasks including link prediction, vertex classification, and commu- nity detection. I am studying NLP problems, including syntactic parsing, machine translation and so on. My research focuses on typical NLP tasks, including word segmentation, tagging and syntactic parsing. I am a NLP researcher in machine translation, especially using deep learning models to improve machine translation. Figure 1: Example of a text-based information network. (Red, blue and green fonts represent con- cerns of the left user, right user and both respec- tively.) In real-world social networks, it is intuitive that one vertex may demonstrate various aspects when interacting with different neighbor vertices. For example, a researcher usually collaborates with various partners on diverse research topics (as il- lustrated in Fig. 1), a social-media user contacts with various friends sharing distinct interests, and a …

Content truncated to 500 words