Matches in SemOpenAlex for { <https://semopenalex.org/work/W2887377541> ?p ?o ?g. }
- W2887377541 abstract "Word2vec embeddings are limited to computing vectors for in-vocabulary terms and do not take into account sub-word information. Character-based representations, such as fastText, mitigate such limitations. We optimize and compare these representations for the biomedical domain. fastText was found to consistently outperform word2vec in named entity recognition tasks for entities such as chemicals and genes. This is likely due to gained information from computed out-of-vocabulary term vectors, as well as the word compositionality of such entities. Contrastingly, performance varied on intrinsic datasets. Optimal hyper-parameters were intrinsic dataset-dependent, likely due to differences in term types distributions. This indicates embeddings should be chosen based on the task at hand. We therefore provide a number of optimized hyper-parameter sets and pre-trained word2vec and fastText models, available on https://github.com/dterg/bionlp-embed." @default.
- W2887377541 created "2018-08-22" @default.
- W2887377541 creator A5011362662 @default.
- W2887377541 creator A5050041550 @default.
- W2887377541 creator A5061524521 @default.
- W2887377541 date "2018-01-01" @default.
- W2887377541 modified "2023-09-24" @default.
- W2887377541 title "Sub-word information in pre-trained biomedical word representations: evaluation and hyper-parameter optimization" @default.
- W2887377541 cites W1832693441 @default.
- W2887377541 cites W1905278307 @default.
- W2887377541 cites W1971220772 @default.
- W2887377541 cites W2037382435 @default.
- W2887377541 cites W2047782770 @default.
- W2887377541 cites W2136480620 @default.
- W2887377541 cites W2145870108 @default.
- W2887377541 cites W2153579005 @default.
- W2887377541 cites W2159092541 @default.
- W2887377541 cites W2250539671 @default.
- W2887377541 cites W2296283641 @default.
- W2887377541 cites W2493916176 @default.
- W2887377541 cites W2509406088 @default.
- W2887377541 cites W2515910099 @default.
- W2887377541 cites W2527896214 @default.
- W2887377541 cites W2613831280 @default.
- W2887377541 cites W2783212081 @default.
- W2887377541 cites W2791951321 @default.
- W2887377541 cites W2962681409 @default.
- W2887377541 cites W2962772361 @default.
- W2887377541 cites W2962986031 @default.
- W2887377541 cites W2963419157 @default.
- W2887377541 cites W2963421945 @default.
- W2887377541 doi "https://doi.org/10.18653/v1/w18-2307" @default.
- W2887377541 hasPublicationYear "2018" @default.
- W2887377541 type Work @default.
- W2887377541 sameAs 2887377541 @default.
- W2887377541 citedByCount "3" @default.
- W2887377541 countsByYear W28873775412019 @default.
- W2887377541 countsByYear W28873775412020 @default.
- W2887377541 crossrefType "proceedings-article" @default.
- W2887377541 hasAuthorship W2887377541A5011362662 @default.
- W2887377541 hasAuthorship W2887377541A5050041550 @default.
- W2887377541 hasAuthorship W2887377541A5061524521 @default.
- W2887377541 hasBestOaLocation W28873775411 @default.
- W2887377541 hasConcept C121332964 @default.
- W2887377541 hasConcept C121375916 @default.
- W2887377541 hasConcept C134306372 @default.
- W2887377541 hasConcept C138885662 @default.
- W2887377541 hasConcept C154945302 @default.
- W2887377541 hasConcept C162324750 @default.
- W2887377541 hasConcept C187736073 @default.
- W2887377541 hasConcept C204321447 @default.
- W2887377541 hasConcept C2524010 @default.
- W2887377541 hasConcept C2776461190 @default.
- W2887377541 hasConcept C2777601683 @default.
- W2887377541 hasConcept C2779135771 @default.
- W2887377541 hasConcept C2780451532 @default.
- W2887377541 hasConcept C2780861071 @default.
- W2887377541 hasConcept C33923547 @default.
- W2887377541 hasConcept C36503486 @default.
- W2887377541 hasConcept C41008148 @default.
- W2887377541 hasConcept C41608201 @default.
- W2887377541 hasConcept C41895202 @default.
- W2887377541 hasConcept C61797465 @default.
- W2887377541 hasConcept C62520636 @default.
- W2887377541 hasConcept C90805587 @default.
- W2887377541 hasConceptScore W2887377541C121332964 @default.
- W2887377541 hasConceptScore W2887377541C121375916 @default.
- W2887377541 hasConceptScore W2887377541C134306372 @default.
- W2887377541 hasConceptScore W2887377541C138885662 @default.
- W2887377541 hasConceptScore W2887377541C154945302 @default.
- W2887377541 hasConceptScore W2887377541C162324750 @default.
- W2887377541 hasConceptScore W2887377541C187736073 @default.
- W2887377541 hasConceptScore W2887377541C204321447 @default.
- W2887377541 hasConceptScore W2887377541C2524010 @default.
- W2887377541 hasConceptScore W2887377541C2776461190 @default.
- W2887377541 hasConceptScore W2887377541C2777601683 @default.
- W2887377541 hasConceptScore W2887377541C2779135771 @default.
- W2887377541 hasConceptScore W2887377541C2780451532 @default.
- W2887377541 hasConceptScore W2887377541C2780861071 @default.
- W2887377541 hasConceptScore W2887377541C33923547 @default.
- W2887377541 hasConceptScore W2887377541C36503486 @default.
- W2887377541 hasConceptScore W2887377541C41008148 @default.
- W2887377541 hasConceptScore W2887377541C41608201 @default.
- W2887377541 hasConceptScore W2887377541C41895202 @default.
- W2887377541 hasConceptScore W2887377541C61797465 @default.
- W2887377541 hasConceptScore W2887377541C62520636 @default.
- W2887377541 hasConceptScore W2887377541C90805587 @default.
- W2887377541 hasLocation W28873775411 @default.
- W2887377541 hasOpenAccess W2887377541 @default.
- W2887377541 hasPrimaryLocation W28873775411 @default.
- W2887377541 hasRelatedWork W1914293925 @default.
- W2887377541 hasRelatedWork W2783279235 @default.
- W2887377541 hasRelatedWork W2891578174 @default.
- W2887377541 hasRelatedWork W2897617454 @default.
- W2887377541 hasRelatedWork W2920052239 @default.
- W2887377541 hasRelatedWork W2964279163 @default.
- W2887377541 hasRelatedWork W2969453628 @default.
- W2887377541 hasRelatedWork W2978368133 @default.
- W2887377541 hasRelatedWork W2990098834 @default.
- W2887377541 hasRelatedWork W3030951257 @default.