Matches in SemOpenAlex for { <https://semopenalex.org/work/W2955102523> ?p ?o ?g. }
Showing items 1 to 81 of
81
with 100 items per page.
- W2955102523 abstract "Extreme multi-label text classification (XMC) aims to tag each input text with the most relevant labels from an extremely large label set, such as those that arise in product categorization and e-commerce recommendation. Recently, pretrained language representation models such as BERT achieve remarkable state-of-the-art performance across a wide range of NLP tasks including sentence classification among small label sets (typically fewer than thousands). Indeed, there are several challenges in applying BERT to the XMC problem. The main challenges are: (i) the difficulty of capturing dependencies and correlations among labels, whose features may come from heterogeneous sources, and (ii) the tractability to scale to the extreme label setting as the model size can be very large and scale linearly with the size of the output space. To overcome these challenges, we propose X-BERT, the first feasible attempt to finetune BERT models for a scalable solution to the XMC problem. Specifically, X-BERT leverages both the label and document text to build label representations, which induces semantic label clusters in order to better model label dependencies. At the heart of X-BERT is finetuning BERT models to capture the contextual relations between input text and the induced label clusters. Finally, an ensemble of the different BERT models trained on heterogeneous label clusters leads to our best final model. Empirically, on a Wiki dataset with around 0.5 million labels, X-BERT achieves new state-of-the-art results where the precision@1 reaches 67:80%, a substantial improvement over 32.58%/60.91% of deep learning baseline fastText and competing XMC approach Parabel, respectively. This amounts to a 11.31% relative improvement over Parabel, which is indeed significant since the recent approach SLICE only has 5.53% relative improvement." @default.
- W2955102523 created "2019-07-12" @default.
- W2955102523 creator A5006559148 @default.
- W2955102523 creator A5017721496 @default.
- W2955102523 creator A5023183059 @default.
- W2955102523 creator A5041434096 @default.
- W2955102523 creator A5063459703 @default.
- W2955102523 date "2019-05-07" @default.
- W2955102523 modified "2023-09-27" @default.
- W2955102523 title "X-BERT: eXtreme Multi-label Text Classification with BERT" @default.
- W2955102523 hasPublicationYear "2019" @default.
- W2955102523 type Work @default.
- W2955102523 sameAs 2955102523 @default.
- W2955102523 citedByCount "3" @default.
- W2955102523 countsByYear W29551025232020 @default.
- W2955102523 countsByYear W29551025232021 @default.
- W2955102523 crossrefType "posted-content" @default.
- W2955102523 hasAuthorship W2955102523A5006559148 @default.
- W2955102523 hasAuthorship W2955102523A5017721496 @default.
- W2955102523 hasAuthorship W2955102523A5023183059 @default.
- W2955102523 hasAuthorship W2955102523A5041434096 @default.
- W2955102523 hasAuthorship W2955102523A5063459703 @default.
- W2955102523 hasConcept C119857082 @default.
- W2955102523 hasConcept C137293760 @default.
- W2955102523 hasConcept C154945302 @default.
- W2955102523 hasConcept C177264268 @default.
- W2955102523 hasConcept C17744445 @default.
- W2955102523 hasConcept C199360897 @default.
- W2955102523 hasConcept C199539241 @default.
- W2955102523 hasConcept C204321447 @default.
- W2955102523 hasConcept C2776359362 @default.
- W2955102523 hasConcept C2777530160 @default.
- W2955102523 hasConcept C2986744138 @default.
- W2955102523 hasConcept C41008148 @default.
- W2955102523 hasConcept C48044578 @default.
- W2955102523 hasConcept C77088390 @default.
- W2955102523 hasConcept C94124525 @default.
- W2955102523 hasConcept C94625758 @default.
- W2955102523 hasConceptScore W2955102523C119857082 @default.
- W2955102523 hasConceptScore W2955102523C137293760 @default.
- W2955102523 hasConceptScore W2955102523C154945302 @default.
- W2955102523 hasConceptScore W2955102523C177264268 @default.
- W2955102523 hasConceptScore W2955102523C17744445 @default.
- W2955102523 hasConceptScore W2955102523C199360897 @default.
- W2955102523 hasConceptScore W2955102523C199539241 @default.
- W2955102523 hasConceptScore W2955102523C204321447 @default.
- W2955102523 hasConceptScore W2955102523C2776359362 @default.
- W2955102523 hasConceptScore W2955102523C2777530160 @default.
- W2955102523 hasConceptScore W2955102523C2986744138 @default.
- W2955102523 hasConceptScore W2955102523C41008148 @default.
- W2955102523 hasConceptScore W2955102523C48044578 @default.
- W2955102523 hasConceptScore W2955102523C77088390 @default.
- W2955102523 hasConceptScore W2955102523C94124525 @default.
- W2955102523 hasConceptScore W2955102523C94625758 @default.
- W2955102523 hasLocation W29551025231 @default.
- W2955102523 hasOpenAccess W2955102523 @default.
- W2955102523 hasPrimaryLocation W29551025231 @default.
- W2955102523 hasRelatedWork W1593841847 @default.
- W2955102523 hasRelatedWork W2088718948 @default.
- W2955102523 hasRelatedWork W2520348554 @default.
- W2955102523 hasRelatedWork W2667174781 @default.
- W2955102523 hasRelatedWork W2739996966 @default.
- W2955102523 hasRelatedWork W2796354582 @default.
- W2955102523 hasRelatedWork W2807910559 @default.
- W2955102523 hasRelatedWork W2903612254 @default.
- W2955102523 hasRelatedWork W2906838467 @default.
- W2955102523 hasRelatedWork W2907669945 @default.
- W2955102523 hasRelatedWork W2916025244 @default.
- W2955102523 hasRelatedWork W2948729947 @default.
- W2955102523 hasRelatedWork W2976463833 @default.
- W2955102523 hasRelatedWork W3080994088 @default.
- W2955102523 hasRelatedWork W3090019957 @default.
- W2955102523 hasRelatedWork W3092911421 @default.
- W2955102523 hasRelatedWork W3153655254 @default.
- W2955102523 hasRelatedWork W3165920028 @default.
- W2955102523 hasRelatedWork W3174569599 @default.
- W2955102523 hasRelatedWork W3177232285 @default.
- W2955102523 isParatext "false" @default.
- W2955102523 isRetracted "false" @default.
- W2955102523 magId "2955102523" @default.
- W2955102523 workType "article" @default.