Matches in SemOpenAlex for { <https://semopenalex.org/work/W2952802852> ?p ?o ?g. }
Showing items 1 to 95 of
95
with 100 items per page.
- W2952802852 abstract "Learning a distinct representation for each sense of an ambiguous word could lead to more powerful and fine-grained models of vector-space representations. Yet while `multi-sense' methods have been proposed and tested on artificial word-similarity tasks, we don't know if they improve real natural language understanding tasks. In this paper we introduce a multi-sense embedding model based on Chinese Restaurant Processes that achieves state of the art performance on matching human word similarity judgments, and propose a pipelined architecture for incorporating multi-sense embeddings into language understanding. We then test the performance of our model on part-of-speech tagging, named entity recognition, sentiment analysis, semantic relation identification and semantic relatedness, controlling for embedding dimensionality. We find that multi-sense embeddings do improve performance on some tasks (part-of-speech tagging, semantic relation identification, semantic relatedness) but not on others (named entity recognition, various forms of sentiment analysis). We discuss how these differences may be caused by the different role of word sense information in each of the tasks. The results highlight the importance of testing embedding models in real applications." @default.
- W2952802852 created "2019-06-27" @default.
- W2952802852 creator A5077851706 @default.
- W2952802852 creator A5087088138 @default.
- W2952802852 date "2015-06-02" @default.
- W2952802852 modified "2023-09-23" @default.
- W2952802852 title "Do Multi-Sense Embeddings Improve Natural Language Understanding?" @default.
- W2952802852 doi "https://doi.org/10.48550/arxiv.1506.01070" @default.
- W2952802852 hasPublicationYear "2015" @default.
- W2952802852 type Work @default.
- W2952802852 sameAs 2952802852 @default.
- W2952802852 citedByCount "0" @default.
- W2952802852 crossrefType "posted-content" @default.
- W2952802852 hasAuthorship W2952802852A5077851706 @default.
- W2952802852 hasAuthorship W2952802852A5087088138 @default.
- W2952802852 hasBestOaLocation W29528028521 @default.
- W2952802852 hasConcept C103278499 @default.
- W2952802852 hasConcept C105795698 @default.
- W2952802852 hasConcept C115961682 @default.
- W2952802852 hasConcept C116834253 @default.
- W2952802852 hasConcept C124101348 @default.
- W2952802852 hasConcept C130318100 @default.
- W2952802852 hasConcept C138885662 @default.
- W2952802852 hasConcept C154945302 @default.
- W2952802852 hasConcept C162324750 @default.
- W2952802852 hasConcept C165064840 @default.
- W2952802852 hasConcept C17744445 @default.
- W2952802852 hasConcept C187736073 @default.
- W2952802852 hasConcept C195324797 @default.
- W2952802852 hasConcept C199539241 @default.
- W2952802852 hasConcept C204321447 @default.
- W2952802852 hasConcept C25343380 @default.
- W2952802852 hasConcept C2776359362 @default.
- W2952802852 hasConcept C2777462759 @default.
- W2952802852 hasConcept C2779439875 @default.
- W2952802852 hasConcept C2780451532 @default.
- W2952802852 hasConcept C33923547 @default.
- W2952802852 hasConcept C41008148 @default.
- W2952802852 hasConcept C41608201 @default.
- W2952802852 hasConcept C41895202 @default.
- W2952802852 hasConcept C44572571 @default.
- W2952802852 hasConcept C59822182 @default.
- W2952802852 hasConcept C66402592 @default.
- W2952802852 hasConcept C86803240 @default.
- W2952802852 hasConcept C90805587 @default.
- W2952802852 hasConcept C94625758 @default.
- W2952802852 hasConceptScore W2952802852C103278499 @default.
- W2952802852 hasConceptScore W2952802852C105795698 @default.
- W2952802852 hasConceptScore W2952802852C115961682 @default.
- W2952802852 hasConceptScore W2952802852C116834253 @default.
- W2952802852 hasConceptScore W2952802852C124101348 @default.
- W2952802852 hasConceptScore W2952802852C130318100 @default.
- W2952802852 hasConceptScore W2952802852C138885662 @default.
- W2952802852 hasConceptScore W2952802852C154945302 @default.
- W2952802852 hasConceptScore W2952802852C162324750 @default.
- W2952802852 hasConceptScore W2952802852C165064840 @default.
- W2952802852 hasConceptScore W2952802852C17744445 @default.
- W2952802852 hasConceptScore W2952802852C187736073 @default.
- W2952802852 hasConceptScore W2952802852C195324797 @default.
- W2952802852 hasConceptScore W2952802852C199539241 @default.
- W2952802852 hasConceptScore W2952802852C204321447 @default.
- W2952802852 hasConceptScore W2952802852C25343380 @default.
- W2952802852 hasConceptScore W2952802852C2776359362 @default.
- W2952802852 hasConceptScore W2952802852C2777462759 @default.
- W2952802852 hasConceptScore W2952802852C2779439875 @default.
- W2952802852 hasConceptScore W2952802852C2780451532 @default.
- W2952802852 hasConceptScore W2952802852C33923547 @default.
- W2952802852 hasConceptScore W2952802852C41008148 @default.
- W2952802852 hasConceptScore W2952802852C41608201 @default.
- W2952802852 hasConceptScore W2952802852C41895202 @default.
- W2952802852 hasConceptScore W2952802852C44572571 @default.
- W2952802852 hasConceptScore W2952802852C59822182 @default.
- W2952802852 hasConceptScore W2952802852C66402592 @default.
- W2952802852 hasConceptScore W2952802852C86803240 @default.
- W2952802852 hasConceptScore W2952802852C90805587 @default.
- W2952802852 hasConceptScore W2952802852C94625758 @default.
- W2952802852 hasLocation W29528028521 @default.
- W2952802852 hasLocation W29528028522 @default.
- W2952802852 hasLocation W29528028523 @default.
- W2952802852 hasOpenAccess W2952802852 @default.
- W2952802852 hasPrimaryLocation W29528028521 @default.
- W2952802852 hasRelatedWork W2151108588 @default.
- W2952802852 hasRelatedWork W2602143361 @default.
- W2952802852 hasRelatedWork W26410400 @default.
- W2952802852 hasRelatedWork W2752417314 @default.
- W2952802852 hasRelatedWork W2935713639 @default.
- W2952802852 hasRelatedWork W2952802852 @default.
- W2952802852 hasRelatedWork W3043988921 @default.
- W2952802852 hasRelatedWork W3214579985 @default.
- W2952802852 hasRelatedWork W4200227521 @default.
- W2952802852 hasRelatedWork W4286799877 @default.
- W2952802852 isParatext "false" @default.
- W2952802852 isRetracted "false" @default.
- W2952802852 magId "2952802852" @default.
- W2952802852 workType "article" @default.