Abstract:Distant supervision is widely used to extract relational facts with automatically labeled datasets to reduce high cost of human annotation. However, current distantly supervised methods suffer from the common problems of word-level and sentence-level noises, which come from a large proportion of irrelevant words in a sentence and inaccurate relation labels for numerous sentences. The problems lead to unacceptable precision in relation extraction and are critical for the success of using distant supervision. In this paper, we propose a novel and robust neural approach to deal with both problems by reducing influences of the multi-granularity noises. Three levels of noises from word, sentence until knowledge type are carefully considered in this work. We first initiate a question-answering based relation extractor (QARE) to remove noisy words in a sentence. Then we use multi-focus multi-instance learning (MMIL) to alleviate the effects of sentence-level noise by utilizing wrongly labeled sentences properly. Finally, to enhance our method against all the noises, we initialize parameters in our method with a priori knowledge learned from the relevant task of entity type classification by transfer learning. Extensive experiments on both existing benchmark and an improved larger dataset demonstrate that our proposed approach remarkably achieves new state-of-the-art performance.

Sampler Discriminator 0 / 1 Sentence Encoder Estimator Noisy Instances Sentence Encoder Estimator Informative InstancesNA Instances Sentence Encoder Estimator Unconfident Instances Confident Instances Sentence Encoder Unconfident Instances Confident Instances

Denoising Distant Supervision for Relation Extraction via Instance-Level Adversarial Training.

Distant Supervision for Relation Extraction with Neural Instance Selector.

Rescue Implicit and Long-tail Cases: Nearest Neighbor Relation Extraction

Neural Relation Extraction with Selective Attention over Instances

SENT: Sentence-level Distant Relation Extraction Via Negative Training

Finding Influential Instances for Distantly Supervised Relation Extraction

Are Noisy Sentences Useless for Distant Supervised Relation Extraction?

Normal Vs. Adversarial: Salience-based Analysis of Adversarial Samples for Relation Extraction

Semantic Piecewise Convolutional Neural Network with Adaptive Negative Training for Distantly Supervised Relation Extraction

Clustering-Augmented Multi-instance Learning for Neural Relation Extraction

Denoising Relation Extraction from Document-level Distant Supervision

Denoising Distant Supervision For Relation Extraction With Entropy Weight Method

Noise Reduction Framework for Distantly Supervised Relation Extraction with Human in the Loop

Few Clean Instances Help Denoising Distant Supervision.

Relation Extraction Method Combining Clause Level Distant Supervision and Semi-supervised Ensemble Learning

Denoising Graph Inference Network for Document-Level Relation Extraction

A Noisy Label and Negative Sample Robust Loss Function for DNN-based Distant Supervised Relation Extraction

Noise-Tolerant Supervised Relation Extraction

A Soft-label Method for Noise-tolerant Distantly Supervised Relation Extraction

Robust Neural Relation Extraction Via Multi-Granularity Noises Reduction