Entity Linking via Joint Encoding of Types, Descriptions, and Context

By Nitish Gupta, Sameer Singh, Dan Roth. 2017

225 citations recorded by the source corpus.

For accurate entity linking, we need to capture various information aspects of an entity, such as its description in a KB, contexts in which it is mentioned, and structured knowledge. Additionally, a linking system should work on texts from different domains without requiring domain-specific training data or hand-engineered features. In this work we present a neural, modular entity linking system that learns a unified dense representation for each entity using multiple sources of information, such as its description, contexts around its mentions, and its fine-grained types. We show that the resulting entity linking system is effective at combining these sources, and performs competitively, sometimes out-performing current state-of-the-art systems across datasets, without requiring any domain-specific training data or hand-engineered features. We also show that our model can effectively “embed” entities that are new to the KB, and is able to link its mentions accurately.

Subfields: Artificial Intelligence, Natural Language Processing

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 Natural Language Processing

Citations in the corpus

Learning Dense Representations for Entity Retrieval (2019, background): “These variations and the resulting fragmentation of evaluation is discussed at length by Ling et al. (2015) and Hachey et al. (2013), and partially addressed by attempts to consolidate datasets (Cornolti et al., 2013) and metrics (Usbeck et al., 2015).”
Barack’s Wife Hillary: Using Knowledge Graphs for Fact-Aware Language Modeling (2019, methodology): “Accordingly, we use the neural-el (Gupta et al., 2017) entity linker to identify additional links to Wikidata, and identify coreferences using Stanford CoreNLP2 to cover pronouns, nominals, and other tokens missed by the linker.”
Ultra-Fine Entity Typing (2018, background): “Fine-grained NER has received growing attention, and is used in many applications (Gupta et al., 2017; Ren et al., 2017; Yaghoobzadeh et al., 2017b; Raiman and Raiman, 2018).”
Joint Entity Linking with Deep Reinforcement Learning (2019, background): “Recently, with the popularity of deep learning models, representation learning is utilized to automatically extract semantic features [2, 17].”
Fine-Grained Entity Typing for Domain Independent Entity Linking (2019, methodology): “This represents a much larger-scale tagset than past work using entity typing for entity linking, which has usually used hundreds of types (Gupta, Singh, and Roth 2017, Murty et al. 2018, Raiman and Raiman 2018).”
Neural Compositional Denotational Semantics for Question Answering (2018, methodology): “To model the large number of entities in KGs such as Freebase, techniques proposed by recent work (Verga et al., 2017; Gupta et al., 2017) that explore representing entities as composition of its properties, such as, types, description etc.”
Learning Cross-Context Entity Representations from Text (2020, background): “…work that makes use of neural context and entity encoders in downstream entity linking systems (Sun et al., 2015; Yamada et al., 2016; 2017; Gupta et al., 2017; Murty et al., 2018; Kolitsas et al., 2018), but that previous work focuses solely on discriminating between entities that match a…”
Neural Cross-Lingual Entity Linking (2017, methodology): “% Hoffart et al. (2011) 82.5 Gupta et al. (2017) 82.9 He et al. (2013) 85.6 Francis-Landau et al. (2016”
Zero-Shot Entity Linking by Reading Entity Descriptions (2019, methodology): “Single vector representations for the two components have been used in many prior works, e.g., Gupta et al. (2017).”
Mention and Entity Description Co-Attention for Entity Disambiguation (2018, methodology): “(Nitish Gupta and Roth 2017) encodes the mention contexts, entity descriptions, and entity types via a neural network jointly.”