Semantic search helper: A tool based on the use of embeddings in multi-item questionnaires as a harmonization opportunity for merging large datasets – A feasibility study
Abstract Background Recent advances in natural language processing (NLP), particularly in language processing methods, have opened new avenues in semantic data analysis. A promising application of NLP is data harmonization in questionnaire-based cohort studies, where it can be used as an additional...
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Cambridge University Press
2025-01-01
|
| Series: | European Psychiatry |
| Subjects: | |
| Online Access: | https://www.cambridge.org/core/product/identifier/S092493382401808X/type/journal_article |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832593709638615040 |
|---|---|
| author | Karl Gottfried Karina Janson Nathalie E. Holz Olaf Reis Johannes Kornhuber Anna Eichler Tobias Banaschewski Frauke Nees IMAC-Mind Consortium |
| author_facet | Karl Gottfried Karina Janson Nathalie E. Holz Olaf Reis Johannes Kornhuber Anna Eichler Tobias Banaschewski Frauke Nees IMAC-Mind Consortium |
| author_sort | Karl Gottfried |
| collection | DOAJ |
| description | Abstract
Background
Recent advances in natural language processing (NLP), particularly in language processing methods, have opened new avenues in semantic data analysis. A promising application of NLP is data harmonization in questionnaire-based cohort studies, where it can be used as an additional method, specifically when only different instruments are available for one construct as well as for the evaluation of potentially new construct-constellations. The present article therefore explores embedding models’ potential to detect opportunities for semantic harmonization.
Methods
Using models like SBERT and OpenAI’s ADA, we developed a prototype application (“Semantic Search Helper”) to facilitate the harmonization process of detecting semantically similar items within extensive health-related datasets. The approach’s feasibility and applicability were evaluated through a use case analysis involving data from four large cohort studies with heterogeneous data obtained with a different set of instruments for common constructs.
Results
With the prototype, we effectively identified potential harmonization pairs, which significantly reduced manual evaluation efforts. Expert ratings of semantic similarity candidates showed high agreement with model-generated pairs, confirming the validity of our approach.
Conclusions
This study demonstrates the potential of embeddings in matching semantic similarity as a promising add-on tool to assist harmonization processes of multiplex data sets and instruments but with similar content, within and across studies.
|
| format | Article |
| id | doaj-art-f5fc7e1f2ad24fc7bfc9a9be77ea4be8 |
| institution | Kabale University |
| issn | 0924-9338 1778-3585 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Cambridge University Press |
| record_format | Article |
| series | European Psychiatry |
| spelling | doaj-art-f5fc7e1f2ad24fc7bfc9a9be77ea4be82025-01-20T10:29:12ZengCambridge University PressEuropean Psychiatry0924-93381778-35852025-01-016810.1192/j.eurpsy.2024.1808Semantic search helper: A tool based on the use of embeddings in multi-item questionnaires as a harmonization opportunity for merging large datasets – A feasibility studyKarl Gottfried0https://orcid.org/0000-0002-2100-3409Karina Janson1https://orcid.org/0000-0002-0902-0628Nathalie E. Holz2Olaf Reis3https://orcid.org/0000-0001-6480-6431Johannes Kornhuber4https://orcid.org/0000-0002-8096-3987Anna Eichler5https://orcid.org/0000-0001-5584-0961Tobias Banaschewski6https://orcid.org/0000-0003-4595-1144Frauke Nees7https://orcid.org/0000-0002-7796-8234IMAC-Mind ConsortiumInstitute of Applied Medical Informatics, University Hospital Center Hamburg-Eppendorf, Hamburg, GermanyDepartment of Child and Adolescent Psychiatry and Psychotherapy, Central Institute of Mental Health, Medical Faculty Mannheim, University of Heidelberg, Mannheim, Baden-Württemberg, Germany Institute of Medical Psychology and Medical Sociology, University Medical Center Schleswig-Holstein, Kiel University, Preußerstraße 1-9, Kiel, Schleswig-Holstein, GermanyDepartment of Child and Adolescent Psychiatry and Psychotherapy, Central Institute of Mental Health, Medical Faculty Mannheim, University of Heidelberg, Mannheim, Baden-Württemberg, Germany German Center for Mental Health (DZPG), Partnersite Mannheim-Heidelberg-Ulm, GermanyDepartment of Child and Adolescent Psychiatry, Neurology, Psychosomatics and Psychotherapy, Rostock University Medical Centre, Rostock, GermanyDepartment of Psychiatry and Psychotherapy, Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU), Erlangen, GermanyDepartment of Child and Adolescent Mental Health, Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU), Erlangen, GermanyDepartment of Child and Adolescent Psychiatry and Psychotherapy, Central Institute of Mental Health, Medical Faculty Mannheim, University of Heidelberg, Mannheim, Baden-Württemberg, Germany German Center for Mental Health (DZPG), Partnersite Mannheim-Heidelberg-Ulm, GermanyDepartment of Child and Adolescent Psychiatry and Psychotherapy, Central Institute of Mental Health, Medical Faculty Mannheim, University of Heidelberg, Mannheim, Baden-Württemberg, Germany Institute of Medical Psychology and Medical Sociology, University Medical Center Schleswig-Holstein, Kiel University, Preußerstraße 1-9, Kiel, Schleswig-Holstein, GermanyAbstract Background Recent advances in natural language processing (NLP), particularly in language processing methods, have opened new avenues in semantic data analysis. A promising application of NLP is data harmonization in questionnaire-based cohort studies, where it can be used as an additional method, specifically when only different instruments are available for one construct as well as for the evaluation of potentially new construct-constellations. The present article therefore explores embedding models’ potential to detect opportunities for semantic harmonization. Methods Using models like SBERT and OpenAI’s ADA, we developed a prototype application (“Semantic Search Helper”) to facilitate the harmonization process of detecting semantically similar items within extensive health-related datasets. The approach’s feasibility and applicability were evaluated through a use case analysis involving data from four large cohort studies with heterogeneous data obtained with a different set of instruments for common constructs. Results With the prototype, we effectively identified potential harmonization pairs, which significantly reduced manual evaluation efforts. Expert ratings of semantic similarity candidates showed high agreement with model-generated pairs, confirming the validity of our approach. Conclusions This study demonstrates the potential of embeddings in matching semantic similarity as a promising add-on tool to assist harmonization processes of multiplex data sets and instruments but with similar content, within and across studies. https://www.cambridge.org/core/product/identifier/S092493382401808X/type/journal_articlenatural language processingharmonizationsemanticquestionnairesbig data |
| spellingShingle | Karl Gottfried Karina Janson Nathalie E. Holz Olaf Reis Johannes Kornhuber Anna Eichler Tobias Banaschewski Frauke Nees IMAC-Mind Consortium Semantic search helper: A tool based on the use of embeddings in multi-item questionnaires as a harmonization opportunity for merging large datasets – A feasibility study European Psychiatry natural language processing harmonization semantic questionnaires big data |
| title | Semantic search helper: A tool based on the use of embeddings in multi-item questionnaires as a harmonization opportunity for merging large datasets – A feasibility study |
| title_full | Semantic search helper: A tool based on the use of embeddings in multi-item questionnaires as a harmonization opportunity for merging large datasets – A feasibility study |
| title_fullStr | Semantic search helper: A tool based on the use of embeddings in multi-item questionnaires as a harmonization opportunity for merging large datasets – A feasibility study |
| title_full_unstemmed | Semantic search helper: A tool based on the use of embeddings in multi-item questionnaires as a harmonization opportunity for merging large datasets – A feasibility study |
| title_short | Semantic search helper: A tool based on the use of embeddings in multi-item questionnaires as a harmonization opportunity for merging large datasets – A feasibility study |
| title_sort | semantic search helper a tool based on the use of embeddings in multi item questionnaires as a harmonization opportunity for merging large datasets a feasibility study |
| topic | natural language processing harmonization semantic questionnaires big data |
| url | https://www.cambridge.org/core/product/identifier/S092493382401808X/type/journal_article |
| work_keys_str_mv | AT karlgottfried semanticsearchhelperatoolbasedontheuseofembeddingsinmultiitemquestionnairesasaharmonizationopportunityformerginglargedatasetsafeasibilitystudy AT karinajanson semanticsearchhelperatoolbasedontheuseofembeddingsinmultiitemquestionnairesasaharmonizationopportunityformerginglargedatasetsafeasibilitystudy AT nathalieeholz semanticsearchhelperatoolbasedontheuseofembeddingsinmultiitemquestionnairesasaharmonizationopportunityformerginglargedatasetsafeasibilitystudy AT olafreis semanticsearchhelperatoolbasedontheuseofembeddingsinmultiitemquestionnairesasaharmonizationopportunityformerginglargedatasetsafeasibilitystudy AT johanneskornhuber semanticsearchhelperatoolbasedontheuseofembeddingsinmultiitemquestionnairesasaharmonizationopportunityformerginglargedatasetsafeasibilitystudy AT annaeichler semanticsearchhelperatoolbasedontheuseofembeddingsinmultiitemquestionnairesasaharmonizationopportunityformerginglargedatasetsafeasibilitystudy AT tobiasbanaschewski semanticsearchhelperatoolbasedontheuseofembeddingsinmultiitemquestionnairesasaharmonizationopportunityformerginglargedatasetsafeasibilitystudy AT fraukenees semanticsearchhelperatoolbasedontheuseofembeddingsinmultiitemquestionnairesasaharmonizationopportunityformerginglargedatasetsafeasibilitystudy AT imacmindconsortium semanticsearchhelperatoolbasedontheuseofembeddingsinmultiitemquestionnairesasaharmonizationopportunityformerginglargedatasetsafeasibilitystudy |