PNEL: Pointer Network based End-To-End Entity Linking over Knowledge Graphs

@inproceedings{Banerjee2020PNELPN,
  title={PNEL: Pointer Network based End-To-End Entity Linking over Knowledge Graphs},
  author={Debayan Banerjee and Debanjan Chaudhuri and Mohnish Dubey and Jens Lehmann},
  booktitle={International Workshop on the Semantic Web},
  year={2020}
}
Question Answering systems are generally modelled as a pipeline consisting of a sequence of steps. In such a pipeline, Entity Linking (EL) is often the first step. Several EL models first perform span detection and then entity disambiguation. In such models errors from the span detection phase cascade to later steps and result in a drop of overall accuracy. Moreover, lack of gold entity spans in training data is a limiting factor for span detector training. Hence the movement towards end-to-end… 
[PDF] Semantic Reader
6 Citations
Highly Influential Citations
1
Background Citations
4
Methods Citations
3

6 Citations

Zero-shot Entity Linking with Less Data

This paper proposes a neuro-symbolic, multi-task learning approach that achieves superior performance on ZEL task with significantly less training data and shows that this approach achieves significantly higher performance than SOTA models when they are trained with just 0.01% or 1% of the original training data.

Tree-KGQA: An Unsupervised Approach for Question Answering Over Knowledge Graphs

The comprehensive evaluations demonstrate that the unsupervised KGQA approach outperforms other supervised state-of-the-art methods on the WebQSP-WD test set (1.4% increase in F1 score) - without training on the target dataset.

Neural Entity Linking: A Survey of Models based on Deep Learning

This work distills a generic architecture of a neural EL system and discusses its components, such as candidate generation, mention-context encoding, and entity ranking, summarizing prominent methods for each of them.

Survey on English Entity Linking on Wikidata

This survey reveals that current Wikidata-specific Entity Linking datasets do not differ in their annotation scheme from schemes for other knowledge graphs like DBpedia, and shows that the potential for multilingual and time-dependent datasets, naturally suited forWikidata, is not lifted.

On the Capabilities of Pointer Networks for Deep Deductive Reasoning

This paper applies neural pointer networks for conducting reasoning over symbolic knowledge bases to explore the benefits and limitations of encoder-decoder architectures in general and pointer networks in particular for developing accurate, generalizable and robust neuro-symbolic reasoners.

Survey on English Entity Linking on Wikidata: Datasets and approaches

This survey reveals that current Wikidata-specific Entity Linking datasets do not differ in their annotation scheme from schemes for other knowledge graphs like DBpedia, and shows that the potential for multilingual and time-dependent datasets, naturally suited forWikidata, is not lifted.

References

SHOWING 1-10 OF 38 REFERENCES

End-to-End Neural Entity Linking

This work proposes the first neural end-to-end EL system that jointly discovers and links entities in a text document and shows that it significantly outperforms popular systems on the Gerbil platform when enough training data is available.

Mixing Context Granularities for Improved Entity Linking on Question Answering Data across Entity Categories

This work investigates entity linking in the context of question answering task and presents a jointly optimized neural architecture for entity mention detection and entity disambiguation that models the surrounding context on different levels of granularity.

EARL: Joint Entity and Relation Linking for Question Answering over Knowledge Graphs

This paper proposes a framework called EARL, which performs entity linking and relation linking as a joint task, and implements two different solution strategies, which significantly outperform the current state-of-the-art approaches for entity and relations linking.

Semantic Parsing on Freebase from Question-Answer Pairs

This paper trains a semantic parser that scales up to Freebase and outperforms their state-of-the-art parser on the dataset of Cai and Yates (2013), despite not having annotated logical forms.

Query-Driven On-The-Fly Knowledge Base Construction

This work proposes a novel approach to build on-the-fly knowledge bases in a query-driven manner, based on a semantic-graph representation of sentences, by which three key IE tasks are performed, namely named-entity disambiguation, co-reference resolution and relation extraction, in a light-weight and integrated manner.

Old is Gold: Linguistic Driven Approach for Entity and Relation Linking of Short Text

The Falcon approach, a light-weight linguistic approach which effectively maps entities and relations within a short text to its mentions of a background knowledge graph, significantly outperforms state-of-the-art entity and relation linking for short text query inventories.

FALCON 2.0: An Entity and Relation Linking framework over Wikidata

This paper presents Falcon 2.0, a rulebased tool capable of accurately mapping entities and relations in short texts to resources in both DBpedia and Wikidata following the same approach in both cases.

Falcon 2.0: An Entity and Relation Linking Tool over Wikidata

This paper presents Falcon 2.0, the first joint entity and relation linking tool over Wikidata, which outperforms all the existing baselines and demonstrates an online API, which can be run without any technical expertise.

LC-QuAD 2.0: A Large Dataset for Complex Question Answering over Wikidata and DBpedia

This article provides LC-QuAD 2.0 (Large-Scale Complex Question Answering Dataset) with 30,000 questions, their paraphrases and their corresponding SPARQL queries and explains how the dataset was created and the variety of questions available.

S-MART: Novel Tree-based Structured Learning Algorithms Applied to Tweet Entity Linking

S-MART, a tree-based structured learning framework based on multiple additive regression trees, is proposed, especially suitable for handling tasks with dense features, and can be used to learn many different structures under various loss functions.