Uncover the Ground-Truth Relations in Distant Supervision: A Neural Expectation-Maximization Framework

TL;DR

This paper introduces a neural EM framework for relation extraction that effectively handles noisy labels in distant supervision, significantly improving the accuracy of uncovering true relation labels from large text corpora.

Contribution

It presents a novel label-denoising approach combining neural networks with probabilistic modeling to address noisy labels in distant supervision for relation extraction.

Findings

Significant performance improvement over existing methods.

Effective in uncovering ground-truth relation labels.

Robust to noisy label conditions.

Abstract

Distant supervision for relation extraction enables one to effectively acquire structured relations out of very large text corpora with less human efforts. Nevertheless, most of the prior-art models for such tasks assume that the given text can be noisy, but their corresponding labels are clean. Such unrealistic assumption is contradictory with the fact that the given labels are often noisy as well, thus leading to significant performance degradation of those models on real-world data. To cope with this challenge, we propose a novel label-denoising framework that combines neural network with probabilistic modelling, which naturally takes into account the noisy labels during learning. We empirically demonstrate that our approach significantly improves the current art in uncovering the ground-truth relation labels.