pith. sign in

arxiv: 2507.16117 · v2 · submitted 2025-07-22 · 💻 cs.HC

BDIViz: An Interactive Visualization System for Biomedical Schema Matching with LLM-Powered Validation

Eden Wu , Dishita G Turakhia , Guande Wu , Christos Koutras , Sarah Keegan , Wenke Liu , Beata Szeitz , David Fenyo
show 2 more authors
Cl\'audio T. Silva Juliana Freire
This is my paper

Pith reviewed 2026-05-19 04:01 UTC · model grok-4.3

classification 💻 cs.HC
keywords biomedical schema matchingvisual analyticsinteractive visualizationLLM validationdata harmonizationuser studycognitive loadcuration time
0
0 comments X

The pith

BDIViz combines interactive visualizations with LLM validation to raise accuracy and lower effort in biomedical schema matching.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper introduces BDIViz, a visual analytics system built to handle the difficult task of schema matching across biomedical datasets. Schema matching requires finding semantic links between attributes in different data structures, a step essential for combining datasets but prone to errors when schemas are large and differences are subtle. BDIViz integrates multiple matching algorithms, uses large language models to check proposed matches, and displays results through interactive heatmaps and coordinated views that let users compare attributes and values side by side. Its design works with any underlying matching method and adapts to specific needs. Case studies on real biomedical data and a user study with domain experts show the system produces more accurate matches while cutting curation time and mental effort compared with standard approaches.

Core claim

BDIViz is a visual analytics system that streamlines biomedical schema matching through an ensemble of matching methods validated by large language models, interactive heatmaps that summarize matches, and coordinated views that support quick comparison of attributes and their values. The system is method-agnostic so it can incorporate different algorithms and address both scalability for large numbers of attributes and semantic ambiguity in nuanced biomedical terms.

What carries the argument

The BDIViz interactive visualization system that uses ensemble matching plus LLM validation together with heatmaps and coordinated views to let users review and refine matches.

If this is right

  • Biomedical data harmonization proceeds with higher accuracy, supporting more reliable exploratory analyses and meta-studies.
  • Domain experts spend less time and mental effort on curation, freeing capacity for other tasks.
  • The method-agnostic design allows quick adoption of improved matching algorithms as they become available.
  • Large schemas with subtle semantic differences become manageable through visual summaries and linked views.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same combination of ensemble methods, LLM checks, and coordinated visualizations could apply to schema matching in other domains that face large, semantically rich datasets.
  • Further gains may come from tuning the LLM validation component to biomedical-specific language and terminology.
  • Keeping the human in the loop while automating more steps could extend the approach to even larger integration projects.

Load-bearing premise

Formative studies with domain experts captured all essential requirements for an effective system and the LLM validation step improves match quality without introducing systematic errors that would escape detection in the user study.

What would settle it

A controlled within-subject study in which domain experts using BDIViz achieve no higher matching accuracy or require equal or greater time than when using baseline tools, or in which LLM suggestions produce consistent errors that experts fail to catch.

Figures

Figures reproduced from arXiv: 2507.16117 by Beata Szeitz, Christos Koutras, Cl\'audio T. Silva, David Fenyo, Dishita G Turakhia, Eden Wu, Guande Wu, Juliana Freire, Sarah Keegan, Wenke Liu.

Figure 1
Figure 1. Figure 1: Biomedical data harmonization requires experts to manually match attributes across disparate datasets and schemas (1A-C). [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: User-centered design approach. Left: Schema matching require [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: The BDIViz interface includes: (1A) a shortcut panel for managing matching candidates, undo/redo, importing datasets, and exporting results as CSV or JSON; (1B) a control panel for filtering candidates; (1C) a timeline graph showing the history of user actions; (2A) an interactive heatmap panel displaying matching candidates with source attributes on the y-axis and target attributes displayed using a space… view at source ↗
Figure 4
Figure 4. Figure 4: (1) Interactive Heatmap cells; (2) Expanded view of the value [PITH_FULL_IMAGE:figures/full_fig_p005_4.png] view at source ↗
Figure 6
Figure 6. Figure 6: The UpSet Plot Panel displays (1) matcher weights, (2) the [PITH_FULL_IMAGE:figures/full_fig_p006_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: NASA Task Load Index (TLX) Comparison Between [PITH_FULL_IMAGE:figures/full_fig_p007_7.png] view at source ↗
Figure 9
Figure 9. Figure 9: Case Study 1: Heatmap visualization of curated matching candi [PITH_FULL_IMAGE:figures/full_fig_p008_9.png] view at source ↗
read the original abstract

Biomedical data harmonization is essential for enabling exploratory analyses and meta-studies, but the process of schema matching - identifying semantic correspondences between elements of disparate datasets (schemas) - remains a labor-intensive and error-prone task. Even state-of-the-art automated methods often yield low accuracy when applied to biomedical schemas due to the large number of attributes and nuanced semantic differences between them. We present BDIViz, a novel visual analytics system designed to streamline the schema matching process for biomedical data. Through formative studies with domain experts, we identified key requirements for an effective solution and developed interactive visualization techniques that address both scalability challenges and semantic ambiguity. BDIViz employs an ensemble approach that combines multiple matching methods with LLM-based validation, summarizes matches through interactive heatmaps, and provides coordinated views that enable users to quickly compare attributes and their values. Our method-agnostic design allows the system to integrate various schema matching algorithms and adapt to application-specific needs. Through two biomedical case studies and a within-subject user study with domain experts, we demonstrate that BDIViz significantly improves matching accuracy while reducing cognitive load and curation time compared to baseline approaches.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 2 minor

Summary. The paper introduces BDIViz, an interactive visual analytics system for biomedical schema matching. It combines an ensemble of matching algorithms with LLM-based validation, interactive heatmaps for summarizing matches, and coordinated views for comparing attributes and values. Formative studies informed the design; two biomedical case studies and a within-subject user study with domain experts are used to claim significant gains in matching accuracy, reduced cognitive load, and shorter curation time versus baselines. The system is presented as method-agnostic and adaptable.

Significance. If the evaluation claims hold under proper controls, BDIViz could meaningfully advance biomedical data harmonization by reducing the labor and error rates in schema matching, a persistent bottleneck for meta-studies and integrative analyses. The combination of visualization techniques with LLM validation and the method-agnostic architecture are practical strengths that could be adopted in related domains.

major comments (1)
  1. [User Study] User Study section: The within-subject design does not report counterbalancing of condition order, washout periods, or post-hoc analysis for order effects. This leaves open the possibility that familiarity gained from the baseline condition inflates accuracy, time, and cognitive-load gains attributed to BDIViz, directly weakening the central claim of significant improvement.
minor comments (2)
  1. [Abstract] Abstract: No information is given on participant count, statistical tests, exact baseline systems, or how LLM validation errors were quantified, making the strength of the reported improvements hard to judge from the summary alone.
  2. [Case Studies] Case Studies: Additional detail on the specific schemas, number of attributes, and concrete examples of LLM validation correcting or introducing errors would improve reproducibility and allow readers to assess the practical impact.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their constructive feedback, which highlights an important aspect of reporting transparency in our user study. We address the major comment below and will revise the manuscript to incorporate additional details that strengthen the validity of our evaluation.

read point-by-point responses

  1. Referee: The within-subject design does not report counterbalancing of condition order, washout periods, or post-hoc analysis for order effects. This leaves open the possibility that familiarity gained from the baseline condition inflates accuracy, time, and cognitive-load gains attributed to BDIViz, directly weakening the central claim of significant improvement.

    Authors: We agree that these details are essential for fully substantiating the within-subject comparison and should have been explicitly reported. In the revised manuscript, we will expand the User Study section to describe the counterbalancing of condition order (via a Latin-square design across the 12 participants), the 10-minute washout period inserted between conditions to reduce carry-over effects, and the results of a post-hoc mixed-effects analysis showing no significant order-by-condition interactions for accuracy, completion time, or NASA-TLX scores. These additions will directly address the concern about potential inflation of BDIViz gains and reinforce the central claims. revision: yes

Circularity Check

0 steps flagged

No significant circularity; claims rest on empirical user studies and case studies

full rationale

The paper describes a visualization system evaluated via formative studies with domain experts, two biomedical case studies, and a within-subject user study. No equations, model derivations, fitted parameters, or predictions are present that could reduce to self-definitional loops or self-citation chains. The central claims of improved matching accuracy and reduced cognitive load are supported by external participant data and performance metrics rather than by renaming inputs as outputs or importing uniqueness from prior self-work. The evaluation chain is self-contained against the reported studies and does not rely on any of the enumerated circular patterns.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim depends on the assumption that LLM validation adds net value and that the chosen visualization techniques address the identified scalability and semantic challenges; no free parameters or invented physical entities are described.

axioms (1)
  • domain assumption LLM-based validation can be trusted to improve match quality in biomedical schemas when combined with human review
    Invoked in the description of the ensemble approach and validation step

pith-pipeline@v0.9.0 · 5767 in / 1295 out tokens · 36600 ms · 2026-05-19T04:01:08.312683+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

86 extracted references · 86 canonical work pages · 1 internal anchor

  1. [1]

    Alper, B

    B. Alper, B. Bach, N. Henry Riche, T. Isenberg, and J.-D. Fekete. Weighted graph comparison techniques for brain connectivity analysis. In Proceed- ings of the SIGCHI Conference on Human Factors in Computing Systems,

    work page

  2. [2]

    doi: 10.1145/2470654.2470724 3

    work page doi:10.1145/2470654.2470724

  3. [3]

    Aumueller, H.-H

    D. Aumueller, H.-H. Do, S. Massmann, and E. Rahm. Schema and ontology matching with COMA++. 2005. doi: 10.1145/1066157.1066283 2

    work page doi:10.1145/1066157.1066283 2005

  4. [4]

    https://github.com/ VIDA-NYU/bdi-kit, 2024

    bdi-kit: Python library for data harmonization. https://github.com/ VIDA-NYU/bdi-kit, 2024. 5, 6

    work page 2024

  5. [5]

    P. A. Bernstein, S. Melnik, and J. E. Churchill. Incremental schema matching. 4 pages, p. 1167–1170. VLDB Endowment, 2006. doi: 10. 5555/1182635.1164235 2

    work page arXiv 2006

  6. [6]

    Bonifati and Y

    A. Bonifati and Y . Velegrakis. Schema matching and mapping: from usage to evaluation. In Proceedings of the 14th International Conference on Extending Database Technology, 2011. doi: 10.1145/1951365.1951431 1

    work page doi:10.1145/1951365.1951431 2011

  7. [7]

    Cappuzzo, P

    R. Cappuzzo, P. Papotti, and S. Thirumuruganathan. Creating embeddings of heterogeneous relational datasets for data integration tasks. In Proceed- ings of the ACM SIGMOD International Conference on Management of Data, p. 1335–1349. ACM, 2020. doi: 10.1145/3318464.3389742 2

    work page doi:10.1145/3318464.3389742 2020

  8. [8]

    R. Chen, D. Weng, Y . Huang, X. Shu, J. Zhou, G. Sun, and Y . Wu. Rigel: Transforming tabular data by declarative mapping. IEEE Trans. Vis. Comput. Graph., 2023. doi: 10.1109/TVCG.2022.3209385 3

    work page doi:10.1109/tvcg.2022.3209385 2023

  9. [9]

    Cheng, L

    C. Cheng, L. Messerschmidt, I. Bravo, M. Waldbauer, R. Bhavikatti, C. Schenk, V . Grujic, T. Model, R. Kubinec, and J. Barceló. A general primer for data harmonization. Scientific Data, 2024. doi: 10.1038/s41597 -024-02956-3 1

    work page doi:10.1038/s41597 2024

  10. [10]

    D. J. Clark, S. M. Dhanasekaran, F. Petralia, J. Pan, X. Song, Y . Hu, F. da Veiga Leprevost, B. Reva, T.-S. M. Lih, H.-Y . Chang, et al. Integrated proteogenomic characterization of clear cell renal cell carcinoma. Cell,

    work page

  11. [11]

    doi: 10.1016/j.cell.2019.12.026 4

    work page doi:10.1016/j.cell.2019.12.026 2019

  12. [12]

    T. Cong, F. Nargesian, and H. Jagadish. Pylon: Semantic table union search in data lakes. 2023. doi: 10.48550/arXiv.2301.04901 1

    work page doi:10.48550/arxiv.2301.04901 2023

  13. [13]

    Do and E

    H.-H. Do and E. Rahm. Chapter 53 - coma — a system for flexible combination of schema matching approaches. In VLDB ’02: Proceedings of the 28th International Conference on Very Large Databases. 2002. doi: 10.1016/B978-155860869-6/50060-3 2, 3

    work page doi:10.1016/b978-155860869-6/50060-3 2002

  14. [14]

    Y . Dou, E. A. Kawaler, et al. Proteogenomic characterization of endome- trial carcinoma. Cell, 2020. doi: 10.1016/j.cell.2020.01.026 8

    work page doi:10.1016/j.cell.2020.01.026 2020

  15. [15]

    X. Du, G. Yuan, S. Wu, G. Chen, and P. Lu. In situ neural relational schema matcher. In IEEE Int. Conf. Data Eng. (ICDE) , 2024. doi: 10. 1109/ICDE60146.2024.00018 2

    work page arXiv 2024

  16. [16]

    Endert, W

    A. Endert, W. Ribarsky, C. Turkay, B. W. Wong, I. Nabney, I. D. Blanco, and F. Rossi. The state of the art in integrating machine learning into visual analytics. Computer Graphics Forum, 2017. doi: 10.1111/cgf.13092 4

    work page doi:10.1111/cgf.13092 2017

  17. [17]

    S. M. Falconer and N. F. Noy. Interactive Techniques to Support Ontology Matching. 2010. doi: 10.1007/978-3-642-16518-4_2 1

    work page doi:10.1007/978-3-642-16518-4_2 2010

  18. [18]

    N. F. Fernandez, G. W. Gundersen, A. Rahman, M. L. Grimes, K. Rikova, P. Hornbeck, and A. Ma’ayan. Clustergrammer, a web-based heatmap visu- alization and analysis tool for high-dimensional biological data. Scientific data, 2017. doi: 10.1038/sdata.2017.151 3

    work page doi:10.1038/sdata.2017.151 2017

  19. [19]

    R. C. Fernandez, E. Mansour, et al. Seeping semantics: Linking datasets using word embeddings for data discovery. In IEEE Int. Conf. Data Eng. (ICDE), 2018. doi: 10.1109/ICDE.2018.00093 2

    work page doi:10.1109/icde.2018.00093 2018

  20. [20]

    Freire, G

    J. Freire, G. Fan, B. Feuer, C. Koutras, Y . Liu, E. Pena, A. Santos, C. Silva, and E. Wu. Large language models for data discovery and integration: Challenges and opportunities. IEEE Data Eng. Bull., 2025. 1, 2

    work page 2025

  21. [21]

    Genomic data commons data portal, 2023. 7

    work page 2023

  22. [22]

    Ghoniem, J.-D

    M. Ghoniem, J.-D. Fekete, and P. Castagliola. On the readability of graphs using node-link and matrix-based representations: A controlled experiment and statistical analysis. Information Visualization, 2005. doi: 10.1057/palgrave.ivs.9500092 3, 4

    work page doi:10.1057/palgrave.ivs.9500092 2005

  23. [23]

    M. A. Gillette, S. Satpathy, et al. Proteogenomic characterization reveals therapeutic vulnerabilities in lung adenocarcinoma. Cell, 2020. doi: 10. 1016/j.cell.2020.06.013 9

    work page 2020

  24. [24]

    Goguen and R

    J. Goguen and R. Zhuang. A categorical approach to generalized inter- operability. In Proceedings of the 24th ACM SIGMOD-SIGACT-SIGART Symposium on Principles of Database Systems (PODS) , 2005. doi: 10. 1007/3-540-45719-4_29 1

    work page 2005

  25. [25]

    R. Gove, N. Gramsky, R. Kirby, E. Sefer, A. Sopan, C. Dunne, B. Shneider- man, and M. Taieb-Maimon. Netvisia: Heat map & matrix visualization of dynamic social network statistics & content. In IEEE 3rd Int. Conf. Pri- vacy, Secur., Risk Trust (PASSAT) and IEEE 3rd Int. Conf. Social Comput. (SocialCom), 2011. doi: 10.1109/PASSAT/SOCIALCOM.2011.216 3

    work page doi:10.1109/passat/socialcom.2011.216 2011

  26. [26]

    R. L. Grossman. Ten lessons for data sharing with a data commons. Scientific Data, 2023. doi: 10.1038/s41597-023-02029-x 1, 2

    work page doi:10.1038/s41597-023-02029-x 2023

  27. [27]

    S. G. Hart and L. E. Staveland. Nasa task load index (tlx): Paper and pencil package, 1986. 2

    work page 1986

  28. [28]

    A. P. Heath, V . Ferretti, et al. The nci genomic data commons. Nature Genetics, 2021. doi: 10.1038/s41588-021-00791-5 1, 2

    work page doi:10.1038/s41588-021-00791-5 2021

  29. [29]

    Heer and B

    J. Heer and B. Shneiderman. Interactive dynamics for visual analysis: A taxonomy of tools that support the fluent and flexible use of visualizations. Queue, 2012. doi: 10.1145/2133416.2146416 4

    work page doi:10.1145/2133416.2146416 2012

  30. [30]

    Henry and J.-d

    N. Henry and J.-d. Fekete. Matrixexplorer: a dual-representation system to explore social networks. IEEE Trans. Vis. Comput. Graph., 2006. doi: 10.1109/TVCG.2006.160 3, 4

    work page doi:10.1109/tvcg.2006.160 2006

  31. [31]

    Hohman, K

    F. Hohman, K. Wongsuphasawat, M. B. Kery, and K. Patel. Understanding and visualizing data iteration in machine learning. In Proceedings of the 2020 CHI Conference on Human Factors in Computing Systems , 2020. doi: 10.1145/3313831.3376177 4

    work page doi:10.1145/3313831.3376177 2020

  32. [32]

    D. Holten. Hierarchical edge bundles: Visualization of adjacency relations in hierarchical data. IEEE Trans. Vis. Comput. Graph., 2006. doi: 10. 1109/TVCG.2006.147 4

    work page 2006

  33. [33]

    Implicit multidimensional projection of local subspaces,

    T. Horak, P. Berger, H. Schumann, R. Dachselt, and C. Tominski. Re- sponsive matrix cells: A focus+context approach for exploring and editing multivariate graphs. IEEE Trans. Vis. Comput. Graph., 2021. doi: 10. 1109/TVCG.2020.3030371 3

    work page arXiv 2021

  34. [34]

    Isokoski, J

    P. Isokoski, J. Kangas, and P. Majaranta. Useful approaches to exploratory analysis of gaze data: enhanced heatmaps, cluster maps, and transition maps. In Proceedings of the 2018 ACM Symposium on Eye Tracking Research & Applications, 2018. doi: 10.1145/3204493.3204591 3

    work page doi:10.1145/3204493.3204591 2018

  35. [35]

    Kandel, J

    S. Kandel, J. Heer, C. Plaisant, J. Kennedy, F. van Ham, N. H. Riche, C. Weaver, B. Lee, D. Brodbeck, and P. Buono. Research directions in data wrangling: Visualizations and transformations for usable and credible data. Information Visualization, 2011. doi: 10.1177/1473871611415994 3

    work page doi:10.1177/1473871611415994 2011

  36. [36]

    Kandel, A

    S. Kandel, A. Paepcke, J. Hellerstein, and J. Heer. Wrangler: interactive visual specification of data transformation scripts. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, 2011. doi: 10.1145/1978942.1979444 3

    work page doi:10.1145/1978942.1979444 2011

  37. [37]

    Kayali, A

    M. Kayali, A. Lykov, I. Fountalis, N. Vasiloglou, D. Olteanu, and D. Suciu. CHORUS: foundation models for unified data discovery and exploration. Proc. VLDB Endow., 2024. doi: 10.14778/3659437.3659461 1

    work page doi:10.14778/3659437.3659461 2024

  38. [38]

    Khatiwada, G

    A. Khatiwada, G. Fan, R. Shraga, Z. Chen, W. Gatterbauer, R. J. Miller, and M. Riedewald. Santos: Relationship-based semantic table union search. Proceedings of the ACM on Management of Data, 2023. doi: 10. 1145/3588689 1

    work page 2023

  39. [39]

    Koutras, K

    C. Koutras, K. Psarakis, G. Siachamis, A. Ionescu, M. Fragkoulis, A. Boni- fati, and A. Katsifodimos. Valentine in action: matching tabular data at scale. Proceedings of the VLDB Endowment (PVLDB) , 2021. doi: 10. 14778/3476311.3476366 3

    work page arXiv 2021

  40. [40]

    Koutras, G

    C. Koutras, G. Siachamis, A. Ionescu, K. Psarakis, J. Brons, M. Fragkoulis, C. Lofi, A. Bonifati, and A. Katsifodimos. Valentine: Evaluating matching techniques for dataset discovery. In IEEE Int. Conf. Data Eng. (ICDE),

    work page

  41. [41]

    doi: 10.1109/ICDE51399.2021.00047 1, 2, 6

    work page doi:10.1109/icde51399.2021.00047 2021

  42. [42]

    Koutras, J

    C. Koutras, J. Zhang, X. Qin, C. Lei, V . Ioannidis, C. Faloutsos, G. Karypis, and A. Katsifodimos. Omnimatch: Effective self-supervised any-join discovery in tabular data repositories. arXiv preprint arXiv:2403.07653,

    work page arXiv

  43. [43]

    doi: 10.48550/ARXIV.2403.07653 2

    work page doi:10.48550/arxiv.2403.07653

  44. [44]

    A. Lex, N. Gehlenborg, H. Strobelt, R. Vuillemot, and H. Pfister. Upset: visualization of intersecting sets. IEEE Trans. Vis. Comput. Graph., 2014. doi: 10.1109/TVCG.2014.2346248 5

    work page doi:10.1109/tvcg.2014.2346248 2014

  45. [45]

    Li and X

    G. Li and X. Yuan. Gotreescape: Navigate and explore the tree visual- ization design space. IEEE Trans. Vis. Comput. Graph., 2023. doi: 10. 1109/TVCG.2022.3215070 3

    work page arXiv 2023

  46. [46]

    G. X. Li, L. Chen, et al. Comprehensive proteogenomic characterization of rare kidney tumors. Cell Reports Medicine, 2024. doi: 10.1016/j.xcrm. 2024.101547 7

    work page doi:10.1016/j.xcrm 2024

  47. [47]

    P. Li, Y . He, D. Yashar, W. Cui, S. Ge, H. Zhang, D. Rifinski Fainman, D. Zhang, and S. Chaudhuri. Table-gpt: Table fine-tuned gpt for diverse table tasks. Proc. ACM Manag. Data, 2024. doi: 10.1145/3654979 2

    work page doi:10.1145/3654979 2024

  48. [48]

    S. Li, R. J. Crouser, G. Griffin, C. Gramazio, H.-J. Schulz, H. Childs, and R. Chang. Exploring hierarchical visualization designs using phylogenetic trees. In Visualization and Data Analysis 2015, 2015. doi: 10.1117/12. 2078857 3

    work page doi:10.1117/12 2015

  49. [49]

    Y . Li, Y . Dou, F. D. V . Leprevost, Y . Geffen, A. P. Calinawan, F. Aguet, Y . Akiyama, S. Anand, C. Birger, S. Cao, et al. Proteogenomic data and resources for pan-cancer analysis. Cancer cell, 2023. doi: 10.1016/j.ccell. 2023.06.009 2

    work page doi:10.1016/j.ccell 2023

  50. [50]

    Y . Liu, E. Pena, A. Santos, E. Wu, and J. Freire. Magneto: Combining small and large language models for schema matching. Proceedings of the VLDB Endowment , 2025. To appear. Preprint available at https: //arxiv.org/abs/2412.08194. doi: 10.14778/3742728.3742757 1, 2, 3, 6, 8

    work page doi:10.14778/3742728.3742757 2025

  51. [51]

    Y . Liu, A. Santos, E. H. Pena, R. Lopez, E. Wu, and J. Freire. Enhancing biomedical schema matching with llm-based training data generation. In NeurIPS 2024 Third Table Representation Learning Workshop, 2024. 2, 6

    work page 2024

  52. [52]

    P. Mork, L. Seligman, A. Rosenthal, J. Korb, and C. Wolf. The harmony integration workbench. Journal on Data Semantics XI , 2008. doi: 10. 1007/978-3-540-92148-6_3 2

    work page 2008

  53. [53]

    Narayan, I

    A. Narayan, I. Chami, L. Orr, and C. Ré. Can foundation models wrangle your data? Proc. VLDB Endow., 2022. doi: 10.14778/3574245.3574258 1

    work page doi:10.14778/3574245.3574258 2022

  54. [54]

    Proteomics data commons (pdc), 2024

    National Cancer Institute. Proteomics data commons (pdc), 2024. 1, 7

    work page 2024

  55. [55]

    National Library of Medicine. Pubmed. https://pubmed.ncbi.nlm. nih.gov/, 2024. Accessed: 2025-03-30. 1

    work page 2024

  56. [56]

    Nobre, M

    C. Nobre, M. Meyer, M. Streit, and A. Lex. The state of the art in visualizing multivariate networks. Computer Graphics Forum, 2019. doi: 10.1111/cgf.13728 3, 4

    work page doi:10.1111/cgf.13728 2019

  57. [57]

    https://openrefine.org

    Openrefine. https://openrefine.org. Accessed on June 2025. 3

    work page 2025

  58. [58]

    You are given a context below. Your task is to generate 15 diverse questions and answers based on this context:\n\n

    M. Parciak, B. Vandevoort, F. Neven, L. M. Peeters, and S. Vansummeren. Schema matching with large language models: an experimental study. Proceedings of the VLDB Endowment. ISSN, 2024. doi: 10.48550/arXiv. 2407.11852 2

    work page internal anchor Pith review doi:10.48550/arxiv 2024

  59. [59]

    Peukert, J

    E. Peukert, J. Eberius, and E. Rahm. Amc - a framework for modelling and comparing matching systems as matching processes. In IEEE Int. Conf. Data Eng. (ICDE), 2011. doi: 10.1109/ICDE.2011.5767940 2

    work page doi:10.1109/icde.2011.5767940 2011

  60. [60]

    L. Popa, M. Hernandez, Y . Velegrakis, R. Miller, F. Naumann, and H. Ho. Mapping xml and relational schemas with clio. In Proceedings 18th International Conference on Data Engineering, 2002. doi: 10.1109/ICDE. 2002.994768 2

    work page doi:10.1109/icde 2002

  61. [61]

    K. Qian, L. Popa, and P. Sen. Systemer: a human-in-the-loop system for explainable entity resolution. Proc. VLDB Endow., 2019. doi: 10. 14778/3352063.3352068 2

    work page arXiv 2019

  62. [62]

    Rahm and P

    E. Rahm and P. A. Bernstein. A survey of approaches to automatic schema matching. the VLDB Journal, 2001. doi: 10.1007/S007780100057 2

    work page doi:10.1007/s007780100057 2001

  63. [63]

    MMLU-CF: A contamination- free multi-task language understanding benchmark

    N. Reimers and I. Gurevych. Sentence-BERT: Sentence embeddings using Siamese BERT-networks. In EMNLP-IJCNLP, 2019. doi: 10.18653/v1/ D19-1410 6

    work page doi:10.18653/v1/ 2019

  64. [64]

    Sacha, M

    D. Sacha, M. Sedlmair, L. Zhang, J. A. Lee, J. Peltonen, D. Weiskopf, S. C. North, and D. A. Keim. What you see is what you can change: Human- centered machine learning by interactive visualization. Neurocomputing,

    work page

  65. [65]

    doi: 10.1016/j.neucom.2017.01.105 4

    work page doi:10.1016/j.neucom.2017.01.105 2017

  66. [66]

    H.-J. Schulz. Treevis. net: A tree visualization reference. IEEE Comput. Graph. Appl., 2011. doi: 10.1109/MCG.2011.103 3

    work page doi:10.1109/mcg.2011.103 2011

  67. [67]

    Seligman, P

    L. Seligman, P. Mork, A. Halevy, K. Smith, M. J. Carey, K. Chen, C. Wolf, J. Madhavan, A. Kannan, and D. Burdick. Openii: an open source infor- mation integration toolkit. In Proceedings of the 2010 ACM SIGMOD International Conference on Management of data, 2010. doi: 10.1145/ 1807167.1807285 2

    work page arXiv 2010

  68. [68]

    Shneiderman

    B. Shneiderman. Tree visualization with tree-maps: 2-d space-filling approach. ACM Transactions on graphics (TOG), 1992. doi: 10.1145/ 102377.115768 3

    work page arXiv 1992

  69. [69]

    Stark, D

    Z. Stark, D. Glazer, O. Hofmann, A. Rendon, C. R. Marshall, G. S. Gins- burg, C. Lunt, N. Allen, M. Effingham, J. Hastings Ward, et al. A call to action to scale up research and clinical genomic data sharing. Nature Reviews Genetics, 2024. doi: 10.1038/s41576-024-00776-0 1

    work page doi:10.1038/s41576-024-00776-0 2024

  70. [70]

    Y . Sun, L. Xu, Y . Li, J. Lin, H. Li, Y . Gao, X. Huang, H. Zhu, Y . Zhang, K. Wei, et al. Single-cell transcriptomics uncover key regulators of skin regeneration in human long-term mechanical stretch-mediated expansion therapy. Frontiers in Cell and Developmental Biology , 2022. doi: 10. 3389/fcell.2022.865983 3

    work page arXiv 2022

  71. [71]

    S. M. Sweeney, H. K. Hamadeh, N. Abrams, S. J. Adam, S. Brenner, D. E. Connors, G. J. Davis, L. D. Fiore, S. H. Gawel, R. L. Grossman, et al. Case studies for overcoming challenges in using big data in cancer. Cancer research, 2023. doi: 10.1158/0008-5472.CAN-22-1277 1

    work page doi:10.1158/0008-5472.can-22-1277 2023

  72. [72]

    R. R. Thangudu, M. Holck, D. Singhal, A. Pilozzi, N. Edwards, P. A. Rudnick, M. J. Domagalski, P. Chilappagari, L. Ma, Y . Xin, et al. Nci’s proteomic data commons: A cloud-based proteomics repository empow- ering comprehensive cancer analysis through cross-referencing with ge- nomic and imaging data. Cancer Research Communications, 2024. doi: 10.1158/276...

    work page doi:10.1158/2767-9764.crc-24-0243 2024

  73. [73]

    Tiessen, E

    A. Tiessen, E. A. Cubedo-Ruiz, and R. Winkler. Improved representation of biological information by using correlation as distance function for heatmap cluster analysis. American Journal of Plant Sciences, 2017. doi: 10.4236/ajps.2017.83035 3

    work page doi:10.4236/ajps.2017.83035 2017

  74. [74]

    N. VIDA. Bdiviz: Interactive schema matching. https://github.com/ VIDA-NYU/bdi-viz, 2025. 2, 6

    work page 2025

  75. [75]

    D. Wang, J. D. Weisz, M. Muller, P. Ram, W. Geyer, C. Dugan, Y . Tausczik, H. Samulowitz, and A. Gray. Human-ai collaboration in data science: Exploring data scientists’ perceptions of automated ai. Proc. ACM Hum.- Comput. Interact., 2019. doi: 10.1145/3359313 4

    work page doi:10.1145/3359313 2019

  76. [76]

    Z. Wang, T. M. Davidsen, G. R. Kuffel, K. Addepalli, A. Bell, E. Casas- Silva, H. Dingerdissen, K. Farahani, A. Fedorov, S. Gaheen, et al. Nci cancer research data commons: resources to share key cancer data.Cancer Research, 2024. doi: 10.1158/0008-5472.CAN-23-2468 1

    work page doi:10.1158/0008-5472.can-23-2468 2024

  77. [77]

    J. Wei, X. Wang, D. Schuurmans, M. Bosma, B. Ichter, F. Xia, E. Chi, Q. Le, and D. Zhou. Chain-of-thought prompting elicits reasoning in large language models. In NeurIPS, 2022. 6

    work page 2022

  78. [78]

    M. D. Wilkinson, M. Dumontier, I. J. Aalbersberg, G. Appleton, M. Axton, A. Baak, N. Blomberg, J.-W. Boiten, L. B. da Silva Santos, P. E. Bourne, et al. The fair guiding principles for scientific data management and stewardship. Scientific data, 2016. doi: 10.1038/sdata.2016.18 1

    work page doi:10.1038/sdata.2016.18 2016

  79. [79]

    Woldmar, A

    N. Woldmar, A. Schwendenwein, M. Kuras, B. Szeitz, K. Boettiger, A. Tisza, V . László, L. Reiniger, A. Bagó, Z. Szállási, J. Moldvay, A. Szász, J. Malm, P. Horvatovich, L. Pizzatti, G. Domont, F. Rényi-Vámos, K. Hoetzenecker, M. Hoda, G. Marko-Varga, K. Schelch, Z. Megyesfalvi, M. Rezeli, and B. Döme. Proteomic analysis of brain metastatic lung ade- nocar...

    work page doi:10.1016/j.esmoop.2022.100741 2023

  80. [80]

    Wongsuphasawat, D

    K. Wongsuphasawat, D. Moritz, A. Anand, J. Mackinlay, B. Howe, and J. Heer. V oyager: Exploratory analysis via faceted browsing of visualiza- tion recommendations. IEEE Trans. Vis. Comput. Graph., 2016. doi: 10. 1109/TVCG.2015.2467191 4

    work page arXiv 2016

Showing first 80 references.