Abstract
Distant supervision is an effective method to generate large scale labeled data for relation extraction, which assumes that if a pair of entities appears in some relation of a Knowledge Graph (KG), all sentences containing those entities in a large unlabeled corpus are then labeled with that relation to train a relation classifier. However, when the pair of entities has multiple relationships in the KG, this assumption may produce noisy relation labels. This paper proposes a label-free distant supervision method, which makes no use of the relation labels under this inadequate assumption, but only uses the prior knowledge derived from the KG to supervise the learning of the classifier directly and softly. Specifically, we make use of the type information and the translation law derived from typical KG embedding model to learn embeddings for certain sentence patterns. As the supervision signal is only determined by the two aligned entities, neither hard relation labels nor extra noise-reduction model for the bag of sentences is needed in this way. The experiments show that the approach performs well in current distant supervision dataset.
Highlights
Distant Supervision was first proposed by Mintz (2009), which used seed triples in Freebase instead of manual annotation to supervise text
It indicates that our label-free supervision with prior knowledge introduced by the translation laws and entity types in Knowledge Graph (KG) is effective in avoiding noise, which can answer the second question we proposed at section 4 credibly
We argue that the noise label problem in distant supervision is mainly caused by the incomplete use of KG information
Summary
Distant Supervision was first proposed by Mintz (2009), which used seed triples in Freebase instead of manual annotation to supervise text It marked text as relation r if (h, r, t) can be found in a known KG, where (h, t) is the pair of entities contained in the text. One way named Multi-Instance Learning(MIL) divided the sentences into different bags by (h, t), and tried to select well-labeled sentences from each bag (Zeng et al, 2015) or reduced the weight of mislabeled data (Lin et al, 2016). As both (T urkey Ankara) and (M exico Guadalajara) will be used to supervise the learning of the encoder for the pattern “in A, B”, it makes the embedding of the sentence pattern closer to the correct relation “/location/location/contains” instead of the wrong relation “/location/country/capital” In this way, we do not need to label the sentences with the hard relation labels anymore. In the experiments, we show that the labelfree approach performs well in current distant supervision dataset
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call