Structured Minimally Supervised Learning for Neural Relation Extraction
This work addresses the challenge of label noise in distant supervision for relation extraction, which is important for natural language processing applications like information extraction.
The paper tackles the problem of minimally supervised relation extraction by combining learned representations with structured learning to predict sentence-level relation mentions using only proposition-level supervision from a knowledge base, achieving state-of-the-art results and outperforming competitive baselines.
We present an approach to minimally supervised relation extraction that combines the benefits of learned representations and structured learning, and accurately predicts sentence-level relation mentions given only proposition-level supervision from a KB. By explicitly reasoning about missing data during learning, our approach enables large-scale training of 1D convolutional neural networks while mitigating the issue of label noise inherent in distant supervision. Our approach achieves state-of-the-art results on minimally supervised sentential relation extraction, outperforming a number of baselines, including a competitive approach that uses the attention layer of a purely neural model.