Fine-Grained Entity Recognition

Ling, Xiao; Weld, Daniel S.

doi:10.1609/aaai.v26i1.8122

Cited by 273 publications

(90 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…this kind of NER tasks are called coarse-grained NER [11,12]. In contrast, the NER tasks whose categories of named entities are much larger are called fine-grained NER [13][14][15][16].…”

Section: Named Entity Recognition (Ner)mentioning

confidence: 99%

SSAR-GNN: Self-Supervised Artist Recommendation with Graph Neural Networks

Wang

Rao

Chen

et al. 2022

Preprint

View full text Add to dashboard Cite

Artist recommendation plays a vital role in the artist domain. Accurate recommendation can help avoid ineffective searches and acquire comprehensive knowledge regarding relationships among artists. However, existing studies mainly focus on artists themselves or artistic works. They are incapable of exploring the relationships among artists in an effective way. In this paper, we study the problem of artist recommendation for the first time. We propose a artist dataset to analyze the similarity relationship from spatial and temporal aspects between artists. Specifically, based on the dataset, we propose a self-supervised learning approach to construct the artist graph. To incorporate the learned graph into existing models, we propose a novel network, SSAR-GNN for recommendation. SSAR-GNN applies a simplified Graph Convolution Network (GCN) on the artist graph to enrich the representation of each artist. Experimental results on the dataset show the effectiveness of our proposed method SSAR-GNN in terms of accuracy.

show abstract

“…this kind of NER tasks are called coarse-grained NER [11,12]. In contrast, the NER tasks whose categories of named entities are much larger are called fine-grained NER [13][14][15][16].…”

Section: Named Entity Recognition (Ner)mentioning

confidence: 99%

SSAR-GNN: Self-Supervised Artist Recommendation with Graph Neural Networks

Wang

Rao

Chen

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…The definition of the label space L varies across literature: some focused on predicting free-form words (e.g., place, hotel, etc) (Ling and Weld, 2012;Choi et al, 2018), while the others aimed at predicting labels organized into a type hierarchy (e.g., /location/hotel) to allow domain knowledge incorporation and ensure label consistency (Gillick et al, 2014). Considering the unclear hierarchy in the studied large-scale label space, we adopt the freeform types without ontologies in this work.…”

Section: Problem Definitionmentioning

confidence: 99%

Does Your Model Classify Entities Reasonably? Diagnosing and Mitigating Spurious Correlations in Entity Typing

Xu¹,

Wang²,

Li³

et al. 2022

Preprint

View full text Add to dashboard Cite

The entity typing task aims at predicting one or more words or phrases that describe the type(s) of a specific mention in a sentence. Due to shortcuts from surface patterns to annotated entity labels and biased training, existing entity typing models are subject to the problem of spurious correlations. To comprehensively investigate the faithfulness and reliability of entity typing methods, we first systematically define distinct kinds of model biases that are reflected mainly from spurious correlations. Particularly, we identify six types of existing model biases, including mentioncontext bias, lexical overlapping bias, named entity bias, pronoun bias, dependency bias, and overgeneralization bias. To mitigate these model biases, we then introduce a counterfactual data augmentation method. By augmenting the original training set with their bias-free counterparts, models are forced to fully comprehend the sentences and discover the fundamental cues for entity typing, rather than relying on spurious correlations for shortcuts. Experimental results on the UFET dataset show that our counterfactual data augmentation approach helps improve generalization of different entity typing models with consistently better performance on both in-and out-ofdistribution test sets.

show abstract

“…Different from a traditional entity-typing task that typically classifies entities into coarse-grained types (e.g., person, location, organization ) [ 3 , 4 ], FET aims to assign an entity with more specific types [ 5 , 6 ], which usually follow a hierarchical structure that can provide more semantic information about the entity [ 7 , 8 ], such as /person/politician , /book/author , etc. FET is a significant subtask of named-entity recognition (NER) [ 9 ] for downstream natural language processing (NLP) applications, such as relation extraction [ 10 , 11 ], question answering [ 12 , 13 ], knowledge base population [ 14 ], and recommendation [ 15 , 16 ].…”

Section: Introductionmentioning

confidence: 99%

“…In FET, knowledge graphs (KGs) usually play an important role. For example, given large-scale KGs, FET systems resort to distant supervision [ 10 ] to generate large training corpora [ 9 , 17 , 18 ] (i.e., to label entity mentions in the training corpus with all types associated with the entity in KGs). Although distant supervision can eliminate the high cost in labeling training data with KGs, how to efficiently encode a KG’s typing knowledge into FET model is still underexplored.…”

Section: Introductionmentioning

confidence: 99%

“…as the priority typing candidates of the mention is still informative for inferring the correct type (i.e., /book/author ) that fits the context best, since they can substantially narrow down the optimal list of possible type labels (referred to as the type’s copying vocabulary from here onward). According to the statistics in Table 1 , over 49% and 64% of all the manually annotated mentions’ typing facts in FIGER (GOLD) [ 9 ] and BBN [ 24 ] (testing data) have already been included in their existing KGs, respectively, which inspires us to improve FET by learning rich known entity-typing information from KGs.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Learning from Knowledge Graphs: Neural Fine-Grained Entity Typing with Copy-Generation Networks

Zhang

Han

et al. 2022

Entropy

View full text Add to dashboard Cite

Fine-grained entity typing (FET) aims to identify the semantic type of an entity in a plain text, which is a significant task for downstream natural language processing applications. However, most existing methods neglect rich known typing information about these entities in knowledge graphs. To address this issue, we take advantage of knowledge graphs to improve fine-grained entity typing through the use of a copy mechanism. Specifically, we propose a novel deep neural model called CopyFet for FET via a copy-generation mechanism. CopyFet can integrate two operations: (i) the regular way of making type inference from the whole type set in the generation model; (ii) the new copy mechanism which can identify the semantic type of a mention with reference to the type-copying vocabulary from a knowledge graph in the copy model. Despite its simplicity, this mechanism proves to be powerful since extensive experiments show that CopyFet outperforms state-of-the-art methods in FET on two benchmark datasets (FIGER (GOLD) and BBN). For example, CopyFet achieves the new state-of-the-art score of 76.4% and 83.6% on the accuracy metric in FIGER (GOLD) and BBN, respectively.

show abstract

Fine-Grained Entity Recognition

Cited by 273 publications

References 21 publications

SSAR-GNN: Self-Supervised Artist Recommendation with Graph Neural Networks

SSAR-GNN: Self-Supervised Artist Recommendation with Graph Neural Networks

Does Your Model Classify Entities Reasonably? Diagnosing and Mitigating Spurious Correlations in Entity Typing

Learning from Knowledge Graphs: Neural Fine-Grained Entity Typing with Copy-Generation Networks

Contact Info

Product

Resources

About