Recently few-shot relation classification has drawn much attention. It devotes to addressing the long-tail relation problem by recognizing the relations from few instances. The existing metric learning methods aim to learn the prototype of classes and make prediction according to distances between query and prototypes. However, it is likely to make unreliable predictions due to the text diversity. It is intuitive that the text descriptions of relation and entity can provide auxiliary support evidence for relation classification. In this paper, we propose TD-Proto, which enhances prototypical network with relation and entity descriptions. We design a collaborative attention module to extract beneficial and instructional information of sentence and entity respectively. A gate mechanism is proposed to fuse both information dynamically so as to obtain a knowledge-aware instance. Experimental results demonstrate that our method achieves excellent performance. CCS CONCEPTS • Computing methodologies → Information extraction.

Distant supervision has obtained great progress on relation classification task. However, it still suffers from noisy labeling problem. Different from previous works that underutilize noisy data which inherently characterize the property of classification, in this paper, we propose RCEND, a novel framework to enhance Relation Classification by Exploiting Noisy Data. First, an instance discriminator with reinforcement learning is designed to split the noisy data into correctly labeled data and incorrectly labeled data. Second, we learn a robust relation classifier in semi-supervised learning way, whereby the correctly and incorrectly labeled data are treated as labeled and unlabeled data respectively. The experimental results show that our method outperforms the state-of-the-art models.

No abstract

Knowledge graph embedding models aim to represent entities and relations in continuous low-dimensional vector space, benefiting many research areas such as knowledge graph completion and web searching. However, previous works do not consider controlling information flow, which makes them hard to obtain useful latent information and limits model performance. Specifically, as human beings, predictions are usually made in multiple steps with every step filtering out irrelevant information and targeting at helpful information. In this paper, we first integrate iterative mechanism into knowledge graph embedding and propose a multi-step gated model which utilizes relations as queries to extract useful information from coarse to fine in multiple steps. First gate mechanism is adopted to control information flow by the interaction between entity and relation with multiple steps. Then we repeat the gate cell for several times to refine the information incrementally. Our model achieves state-of-the-art performance on most benchmark datasets compared to strong baselines. Further analyses demonstrate the effectiveness of our model and its scalability on large knowledge graphs.

No abstract