Boosting Entity Linking Performance by Leveraging Unlabeled Documents

By Phong Le, Ivan Titov. 2019

45 citations recorded by the source corpus.

Modern entity linking systems rely on large collections of documents specifically annotated for the task (e.g., AIDA CoNLL). In contrast, we propose an approach which exploits only naturally occurring information: unlabeled documents and Wikipedia. Our approach consists of two stages. First, we construct a high recall list of candidate entities for each mention in an unlabeled document. Second, we use the candidate lists as weak supervision to constrain our document-level entity linking model. The model treats entities as latent variables and, when estimated on a collection of unlabelled texts, learns to choose entities relying both on local context of each mention and on coherence with other entities in the document. The resulting approach rivals fully-supervised state-of-the-art systems on standard test sets. It also approaches their performance in the very challenging setting: when tested on a test set sampled from the data used to estimate the supervised systems. By comparing to Wikipedia-only training of our model, we demonstrate that modeling unlabeled documents is beneficial.

Subfields: Artificial Intelligence, Information Extraction

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Information Extraction

Citations in the corpus

Fine-Grained Entity Typing for Domain Independent Entity Linking (2019, methodology): “Moreover, we only use Wikipedia as a source of training data; some other work like Le and Titov (2019) transductive setting.”
Improving Entity Linking by Modeling Latent Entity Type Information (2020, background): “A line of follow-up work (Le and Titov 2018; 2019a; 2019b) investigate potential improvement solution or other task settings based on that.”