Relation extraction has been widely used to find semantic relations between entities from plain text. Dependency trees provide deeper semantic information for relation extraction. However, existing dependency tree based models adopt pruning strategies that are too aggressive or conservative, leading to insufficient semantic information or excessive noise in relation extraction models. To overcome this issue, we propose the Neural Attentional Relation Extraction Model with Dual Dependency Trees (called DDT-REM), which takes advantage of both the syntactic dependency tree and the semantic dependency tree to well capture syntactic features and semantic features, respectively. Specifically, we first propose novel representation learning to capture the dependency relations from both syntax and semantics. Second, for the syntactic dependency tree, we propose a local-global attention mechanism to solve semantic deficits. We design an extension of graph convolutional networks (GCNs) to perform relation extraction, which effectively improves the extraction accuracy. We conduct experimental studies based on three real-world datasets. Compared with the traditional methods, our method improves the F1 scores by 0.3, 0.1 and 1.6 on three real-world datasets, respectively.

Identifying accounts across different online social networks that belong to the same user has attracted extensive attentions. However, existing techniques rely on given user seeds and ignore the dynamic changes of online social networks, which fails to generate high quality identification results. In order to solve this problem, we propose an incremental user identification method based on user-guider similarity index (called CURIOUS), which efficiently identifies users and well captures the changes of user features over time. Specifically, we first construct a novel user-guider similarity index (called USI) to speed up the matching between users. Second we propose a two-phase user identification strategy consisting of USI-based bidirectional user matching and seed-based user matching, which is effective even for incomplete networks. Finally, we propose incremental maintenance for both USI and the identification results, which dynamically captures the instant states of social networks. We conduct experimental studies based on three real-world social networks. The experiments demonstrate the effectiveness and the efficiency of our proposed method in comparison with traditional methods. Compared with the traditional methods, our method improves precision, recall and rank score by an average of 0.19, 0.16 and 0.09 respectively, and reduces the time cost by an average of 81%.