Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386392112> ?p ?o ?g. }
Showing items 1 to 97 of
97
with 100 items per page.
- W4386392112 abstract "Pooling data from different sources may help further mental health research by providing larger sample sizes and allowing cross-study comparisons; however, the heterogeneity in how variables are measured across studies poses a significant challenge to this process. This study explores the potential of natural language processing (NLP) to harmonize different mental health questionnaires by matching similar items based on their semantic content. Using the Sentence-BERT model, the content of 39 questions from 5 mental health scales was converted into numeric vectors representing their semantic content. These vectors enabled the calculation of cosine similarity scores, which served as a measure of the semantic similarity between items (N=741 item pairs). Using data from a representative UK sample of adults (N=2,058), Spearman rank correlations were also calculated for the same pairs of items. We then tested Pearson correlations between these two indices, and found a moderate overall correlation (r = .48, p <.001) between the cosine similarity scores and the Spearman correlation coefficients. In a holdout sample, these cosine scores exhibited the ability to predict real-world correlations with a mean error of +/- 0.05, suggesting the utility of NLP in identifying similar items for cross-study data pooling. This indicates that the semantic similarity score between two questions can predict the actual correlation of how participants would answer the same two items. However, NLP struggled to replicate more complex correlational structures (i.e. latent factors) found in real-world data. This research contributes to the burgeoning field of retrospective data harmonization by highlighting the potential of NLP to facilitate cross-study data pooling in mental health research. Nevertheless, researchers are cautioned to verify the psychometric equivalence of matched items, as NLP may not fully capture intricate data structures." @default.
- W4386392112 created "2023-09-03" @default.
- W4386392112 creator A5000888409 @default.
- W4386392112 creator A5004665500 @default.
- W4386392112 creator A5018468455 @default.
- W4386392112 creator A5034460774 @default.
- W4386392112 creator A5038610271 @default.
- W4386392112 creator A5075817903 @default.
- W4386392112 creator A5076596849 @default.
- W4386392112 creator A5082716112 @default.
- W4386392112 date "2023-09-01" @default.
- W4386392112 modified "2023-09-27" @default.
- W4386392112 title "Using natural language processing to facilitate the harmonization of mental health questionnaires: a validation study using real-world data" @default.
- W4386392112 doi "https://doi.org/10.31234/osf.io/rxpv9" @default.
- W4386392112 hasPublicationYear "2023" @default.
- W4386392112 type Work @default.
- W4386392112 citedByCount "0" @default.
- W4386392112 crossrefType "posted-content" @default.
- W4386392112 hasAuthorship W4386392112A5000888409 @default.
- W4386392112 hasAuthorship W4386392112A5004665500 @default.
- W4386392112 hasAuthorship W4386392112A5018468455 @default.
- W4386392112 hasAuthorship W4386392112A5034460774 @default.
- W4386392112 hasAuthorship W4386392112A5038610271 @default.
- W4386392112 hasAuthorship W4386392112A5075817903 @default.
- W4386392112 hasAuthorship W4386392112A5076596849 @default.
- W4386392112 hasAuthorship W4386392112A5082716112 @default.
- W4386392112 hasBestOaLocation W43863921121 @default.
- W4386392112 hasConcept C101601086 @default.
- W4386392112 hasConcept C103278499 @default.
- W4386392112 hasConcept C105795698 @default.
- W4386392112 hasConcept C114614502 @default.
- W4386392112 hasConcept C115961682 @default.
- W4386392112 hasConcept C117220453 @default.
- W4386392112 hasConcept C118552586 @default.
- W4386392112 hasConcept C119857082 @default.
- W4386392112 hasConcept C129848803 @default.
- W4386392112 hasConcept C130318100 @default.
- W4386392112 hasConcept C134362201 @default.
- W4386392112 hasConcept C153180895 @default.
- W4386392112 hasConcept C154945302 @default.
- W4386392112 hasConcept C15744967 @default.
- W4386392112 hasConcept C164226766 @default.
- W4386392112 hasConcept C165064840 @default.
- W4386392112 hasConcept C185592680 @default.
- W4386392112 hasConcept C198531522 @default.
- W4386392112 hasConcept C204321447 @default.
- W4386392112 hasConcept C2524010 @default.
- W4386392112 hasConcept C2777530160 @default.
- W4386392112 hasConcept C2780762811 @default.
- W4386392112 hasConcept C2781162219 @default.
- W4386392112 hasConcept C33923547 @default.
- W4386392112 hasConcept C41008148 @default.
- W4386392112 hasConcept C43617362 @default.
- W4386392112 hasConcept C70437156 @default.
- W4386392112 hasConceptScore W4386392112C101601086 @default.
- W4386392112 hasConceptScore W4386392112C103278499 @default.
- W4386392112 hasConceptScore W4386392112C105795698 @default.
- W4386392112 hasConceptScore W4386392112C114614502 @default.
- W4386392112 hasConceptScore W4386392112C115961682 @default.
- W4386392112 hasConceptScore W4386392112C117220453 @default.
- W4386392112 hasConceptScore W4386392112C118552586 @default.
- W4386392112 hasConceptScore W4386392112C119857082 @default.
- W4386392112 hasConceptScore W4386392112C129848803 @default.
- W4386392112 hasConceptScore W4386392112C130318100 @default.
- W4386392112 hasConceptScore W4386392112C134362201 @default.
- W4386392112 hasConceptScore W4386392112C153180895 @default.
- W4386392112 hasConceptScore W4386392112C154945302 @default.
- W4386392112 hasConceptScore W4386392112C15744967 @default.
- W4386392112 hasConceptScore W4386392112C164226766 @default.
- W4386392112 hasConceptScore W4386392112C165064840 @default.
- W4386392112 hasConceptScore W4386392112C185592680 @default.
- W4386392112 hasConceptScore W4386392112C198531522 @default.
- W4386392112 hasConceptScore W4386392112C204321447 @default.
- W4386392112 hasConceptScore W4386392112C2524010 @default.
- W4386392112 hasConceptScore W4386392112C2777530160 @default.
- W4386392112 hasConceptScore W4386392112C2780762811 @default.
- W4386392112 hasConceptScore W4386392112C2781162219 @default.
- W4386392112 hasConceptScore W4386392112C33923547 @default.
- W4386392112 hasConceptScore W4386392112C41008148 @default.
- W4386392112 hasConceptScore W4386392112C43617362 @default.
- W4386392112 hasConceptScore W4386392112C70437156 @default.
- W4386392112 hasLocation W43863921121 @default.
- W4386392112 hasOpenAccess W4386392112 @default.
- W4386392112 hasPrimaryLocation W43863921121 @default.
- W4386392112 hasRelatedWork W1020482675 @default.
- W4386392112 hasRelatedWork W1997312918 @default.
- W4386392112 hasRelatedWork W2047828095 @default.
- W4386392112 hasRelatedWork W2116838603 @default.
- W4386392112 hasRelatedWork W2359659532 @default.
- W4386392112 hasRelatedWork W2365659184 @default.
- W4386392112 hasRelatedWork W2766760871 @default.
- W4386392112 hasRelatedWork W3078371441 @default.
- W4386392112 hasRelatedWork W3182591145 @default.
- W4386392112 hasRelatedWork W4226066015 @default.
- W4386392112 isParatext "false" @default.
- W4386392112 isRetracted "false" @default.
- W4386392112 workType "article" @default.