Up to now, relation extraction systems have made extensive use of features generated by linguistic analysis modules. Errors in these features lead to errors of relation detection and classification. In this work, we depart from these traditional approaches with complicated feature engineering by introducing a convolutional neural network for relation extraction that automatically learns features from sentences and minimizes the dependence on external toolkits and resources. Our model takes advantages of multiple window sizes for filters and pre-trained word embeddings as an initializer on a non-static architecture to improve the performance. We emphasize the relation extraction problem with an unbalanced corpus. The experimental results show that our system significantly outperforms not only the best baseline systems for relation extraction but also the state-of-the-art systems for relation classification.

/pdf/relation-extraction-perspective-from-convolutional-neural-4yydbob1ng.pdf

Relation Extraction: Perspective from Convolutional Neural Networks

Neural models have achieved remarkable success on relation extraction (RE) benchmarks. However, there is no clear understanding what information in text affects existing RE models to make decisions and how to further improve the performance of these models. To this end, we empirically study the effect of two main information sources in text: textual context and entity mentions (names). We find that (i) while context is the main source to support the predictions, RE models also heavily rely on the information from entity mentions, most of which is type information, and (ii) existing datasets may leak shallow heuristics via entity mentions and thus contribute to the high performance on RE benchmarks. Based on the analyses, we propose an entity-masked contrastive pre-training framework for RE to gain a deeper understanding on both textual context and type information while avoiding rote memorization of entities or use of superficial cues in mentions. We carry out extensive experiments to support our views, and show that our framework can improve the effectiveness and robustness of neural models in different RE scenarios. All the code and datasets are released at https://github.com/thunlp/RE-Context-or-Names.

Learning from Context or Names? An Empirical Study on Neural Relation Extraction

Relation Extraction is an important sub-task of Information Extraction which has the potential of employing deep learning (DL) models with the creation of large datasets using distant supervision. In this review, we compare the contributions and pitfalls of the various DL models that have been used for the task, to help guide the path ahead.

A Survey of Deep Learning Methods for Relation Extraction.

Chinese relation extraction is conducted using neural networks with either character-based or word-based inputs, and most existing methods typically suffer from segmentation errors and ambiguity of polysemy. To address the issues, we propose a multi-grained lattice framework (MG lattice) for Chinese relation extraction to take advantage of multi-grained language information and external linguistic knowledge. In this framework, (1) we incorporate word-level information into character sequence inputs so that segmentation errors can be avoided. (2) We also model multiple senses of polysemous words with the help of external linguistic knowledge, so as to alleviate polysemy ambiguity. Experiments on three real-world datasets in distinct domains show consistent and significant superiority and robustness of our model, as compared with other baselines. We will release the source code of this paper in the future.

/pdf/chinese-relation-extraction-with-multi-grained-information-52srbhohee.pdf

Chinese Relation Extraction with Multi-Grained Information and External Linguistic Knowledge.

Continual relation learning aims to continually train a model on new data to learn incessantly emerging novel relations while avoiding catastrophically forgetting old relations. Some pioneering work has proved that storing a handful of historical relation examples in episodic memory and replaying them in subsequent training is an effective solution for such a challenging problem. However, these memory-based methods usually suffer from overfitting the few memorized examples of old relations, which may gradually cause inevitable confusion among existing relations. Inspired by the mechanism in human long-term memory formation, we introduce episodic memory activation and reconsolidation (EMAR) to continual relation learning. Every time neural models are activated to learn both new and memorized data, EMAR utilizes relation prototypes for memory reconsolidation exercise to keep a stable understanding of old relations. The experimental results show that EMAR could get rid of catastrophically forgetting old relations and outperform the state-of-the-art continual learning models.

/pdf/continual-relation-learning-via-episodic-memory-activation-26jksui47u.pdf

Continual Relation Learning via Episodic Memory Activation and Reconsolidation

Deep Neural Network has been applied to many Natural Language Processing tasks. Instead of building hand-craft features, DNN builds features by automatic learning, fitting different domains well. In this paper, we propose a novel convolution network, incorporating lexical features, applied to Relation Extraction. Since many current deep neural networks use word embedding by word table, which, however, neglects semantic meaning among words, we import a new coding method, which coding input words by synonym dictionary to integrate semantic knowledge into the neural network. We compared our Convolution Neural Network CNN on relation extraction with the state-of-art tree kernel approach, including Typed Dependency Path Kernel and Shortest Dependency Path Kernel and Context-Sensitive tree kernel, resulting in a 9% improvement competitive performance on ACE2005 data set. Also, we compared the synonym coding with the one-hot coding, and our approach got 1.6% improvement. Moreover, we also tried other coding method, such as hypernym coding, and give some discussion according the result.

Wenbo Sun

Papers

Convolution Neural Network for Relation Extraction