Pith. sign in

REVIEW 3 major objections 5 minor 77 references

Rethinking Dataset Discovery with DataScout

T0 review · 3 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

Pith's one-line read Proactive AI assistance changes how users explore and understand dataset search spaces.

desk verdict DataScout is a solid, honest HCI systems paper with a real contribution, but the success-rate comparison is contaminated by task-specific corpus augmentation — read the qualitative findings, treat the numbers cautiously. read the letter →

arxiv 2507.18971 v1 pith:R2B27J34 submitted 2025-07-25 cs.HC

classification cs.HC
keywords exploratorydatasetsearchdiscoveryLLMassistancequeryreformulationsemanticrelevanceindicatorssensemakinghuman-AIinteraction
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

Dataset search fails because users cannot express half-formed preferences and cannot see what datasets exist or why results appear. DataScout attacks both failures by making the search itself a conversation: it reformulates the user's query into options grounded in what is actually in the result set, offers filters over column meaning and row granularity, and generates per-dataset explanations of utility and limitation for the stated task. In a within-subjects study against keyword and plain semantic search, users of DataScout assessed dataset suitability faster, reached a promising dataset sooner, and succeeded more often. The qualitative finding is that users treated these proactive features as feedback channels, rewriting queries to steer the explanations and using filter suggestions as sanity checks that they were in the right part of the search space.

What carries the argument

The load-bearing mechanism is a hybrid offline and online pipeline. Offline, each table is augmented by an LLM with summaries, column descriptions, source and collection details, granularity tags, and purposes, then embedded three ways: whole dataset, individual attributes, and purpose, with indexes built for dataset and attribute retrieval. Online, the user's query is expanded into three hypothetical table schemas whose embeddings are averaged against the dataset index; the retrieved result set is then clustered over purpose embeddings for reformulation suggestions and over attribute embeddings for column-concept filters, with the most frequent granularity tags offered as filters; and the top results receive LLM-generated utility and limitation indicators conditioned on the query and filters. This makes every suggestion a function of what the search actually returned, which is what lets DataScout ground reformulations in available data rather than open-ended generation.

What would settle it

Take a held-out set of tables, run the same metadata-augmentation and relevance-indicator prompts, and compare every generated utility and limitation statement and granularity tag against the actual rows and a human label; if a material fraction of indicators assert facts not present in, or contradicted by, the preview rows (for example a time range, geographic scope, or "No significant utilities" verdict the data does not support), the trust and sensemaking effects reported in the study would not generalize.

Watch

Extended reading notes

Core claim

On its own terms, the paper claims that proactive, search-space-aware assistance changes how people discover datasets. DataScout generates three hypothetical schemas from the user's task to retrieve semantically relevant datasets, clusters the retrieved datasets by purpose to propose query reformulations that are guaranteed to have matching results, and clusters attributes to propose column-concept filters and granularity filters drawn from the result set. It then produces task-specific utility and limitation statements for inspected datasets. The user study reports that this combination shortened average suitability assessment from about two minutes in the baselines to 37 seconds, raised success from 7 of 12 and 6 of 12 participants to 10 of 12, and most distinctively led participants to treat suggestions as implicit feedback about query quality and dataset availability, building a conceptual model of the search space and even transferring that awareness back to baseline tools.

Load-bearing premise

The system assumes the LLM-generated dataset summaries, granularity tags, and relevance indicators are faithful to what the tables actually contain; if those annotations hallucinate or misdescribe the data, users would be misled and the study's measured benefits would not transfer beyond this corpus.

Editorial extensions

If this is right

  • Relevance indicators double as query feedback: when top-ranked datasets show "No significant utilities," users infer that the search space lacks what they asked for and reformulate, turning a retrieval list into a diagnostic.
  • Attribute-level semantic filters recover datasets that exact column-name matching misses, including for expert users in their own domains; one participant found a relevant art-history collection they had not previously encountered.
  • Users apply the conceptual model learned from DataScout's suggestions even after switching to baseline tools, for example checking dataset upload dates more deliberately, so the sensemaking benefit is not confined to the interface itself.
  • Supporting both broad starting points of 50 or more results and targeted narrowing to 10 to 12 results fits the observed workflow of exploration followed by refinement, rather than a single retrieval mode.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The reported "query as knobs" behavior suggests a design principle: in exploratory search, users actively steer the explanation generator; a testable extension is making the dimensions of relevance indicators explicit and user-selectable to see whether the sensemaking benefits persist.
  • If LLM-generated metadata is the weak link, the same interaction design could be re-run with human-verified or schema-derived annotations; the study does not establish how performance degrades under annotation error.
  • Grounding query suggestions in clusters of actual results is a transferable pattern: code search, document corpora, and other open-ended discovery tasks face the same gap between user intent and available items, and could reuse the cluster-then-reformulate mechanism.
  • The qualitative mechanisms of feedback loops, sanity checks, and conceptual model building could be quantified in a larger study by logging reformulation decisions and measuring post-task transfer to unassisted search.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. This paper introduces DataScout, a dataset search interface that combines semantic search with three LLM-powered proactive assistance features: query reformulation suggestions grounded in the current result space, semantic attribute and granularity filters, and task-specific dataset relevance indicators. The authors motivate the design with a formative study (n=8) and evaluate with a within-subjects user study (n=12) comparing DataScout against Kaggle keyword search and a stripped semantic baseline, reporting quantitative metrics (success counts, time-to-target, ease-of-use ratings) and qualitative themes about steering, sensemaking, and conceptual model building. The paper argues that DataScout measurably improves exploratory dataset discovery and that its features are uniquely used for structured exploration and for gleaning feedback on search queries.

Significance. If the findings hold, DataScout makes a worthwhile contribution to HCI for data science: it operationalizes design considerations for exploratory dataset search and provides qualitative evidence that search-space-aware LLM suggestions function as feedback mechanisms. The paper is transparent about implementation, with appendices containing the prompts for metadata generation and online assistance, and the system description is detailed enough to reproduce. The strengths are the formative study, the within-subjects design, and the rich qualitative analysis with participant quotes. However, the headline quantitative claim of improved task success is currently threatened by a corpus-augmentation confound, and the system's reliance on unvalidated LLM metadata creates a transferability risk.

major comments (3)
  1. [Section 6, Table 4] The comparison underlying the paper's headline success/failure numbers is confounded. The authors state that after indexing 6,500 Kaggle datasets they 'augmented our initial dataset collection by indexing 300 additional datasets, containing top 25 Kaggle dataset search results for each participant's task.' This means DataScout and the semantic baseline searched a corpus explicitly seeded with task-relevant datasets, while the Kaggle keyword baseline (condition A) was the live Kaggle site with no such task-specific seed injection. The success counts (10/12 vs 7/12 and 6/12), datasets explored (6.02 vs 3.33/4.25), and time-to-first-target (5.1 vs 7.0/7.5 min) therefore cannot be identified as effects of DataScout's interface; they may partly reflect the different, task-curated search space. Disclosure of the augmentation in the text does not remove the confound. Please either compare against a keyword baseline over the same augmented corpus, analyze the non-augmented subset separately, or substantially soften the claim that DataScout 'measurably improves' task success; also report whether any difference in success counts is statistically meaningful at this sample size.
  2. [Sections 5.1 and 5.3.4] DataScout's core features are built on LLM-generated metadata (summaries, purposes, column descriptions, granularity tags) and LLM-generated relevance indicators, with no ground-truth verification or error analysis. The prompt in Appendix C instructs the model to 'Stay factual' and 'Avoid hallucination,' but there is no mechanism that checks the generated text against the actual dataset contents; Section 5.3.4 even generates 'utilities' and 'limitations' from descriptions and example rows. If these outputs hallucinate or mischaracterize data, users would be misled, and the positive qualitative findings about relevance indicators would not transfer to other corpora or LLM backends. Please add a validation study (e.g., manual evaluation of a random sample of metadata and indicators, or a comparison against dataset-level ground truth) and discuss how accuracy failures affect the design conclusions.
  3. [Section 7, Table 4] The paper reports quantitative advantages for DataScout—'found relevant datasets sooner' (5.1 vs 7.0/7.5 min) and faster suitability assessment (37s vs 115/134s)—but does not report significance tests or per-participant distributions for these measures, and the success criterion is self-judged 'promising' with no fixed target datasets. With n=12 and three conditions, these numbers could be driven by a few participants or by the corpus difference noted above. Please report the underlying distributions, conduct appropriate paired tests (or justify their absence), and clearly label the success comparison as exploratory rather than confirmatory.
minor comments (5)
  1. [Section 1] The text 'DataScoutassists users' is missing a space between 'DataScout' and 'assists'; please fix the typo.
  2. [Figure 2] The callout labels in Figure 2 are difficult to parse; for example, the label 'A' appears alongside '(11)' and 'B', and the caption does not map all labeled panels to the prose. Please clean up the labels and ensure each panel is referenced consistently in the text.
  3. [Section 7.2.1] The sentence 'We hypothesize that the presence of relevance indicators but facilitate meta-cognition' appears to contain a typo: 'but' should likely be 'may'. Please correct it.
  4. [Section 8.3] The limitation that only two observations per condition order were recorded is acknowledged, but the paper should also state the implication for the order-effect claims in Section 7, where differences in feature appreciation by exposure order are reported without supporting statistical or systematic evidence.
  5. [Section 6] The citation [18] for the within-subjects repeated-measures design appears to be Hearst's 'Search User Interfaces'; consider citing a standard methods or experimental-design reference instead.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the user study and system evaluation are external to the construction, and no prediction reduces to its inputs by definition.

full rationale

The paper contains no derivation chain in which an output is equivalent to an input by construction. DataScout's offline pipeline (Section 5.1) computes embeddings and LLM-generated metadata before the user study, and the online assistance features (Section 5.3) are grounded in search results and user queries but are evaluated through an independent within-subjects study with external task outcomes. The design considerations are motivated partly by the authors' prior survey [22], but that citation is used as background motivation, not as the evidence for the paper's central empirical claims; the formative study (n=8) and user study (n=12) provide independent qualitative and behavioral evidence. No fitted parameter is renamed as a prediction, no uniqueness theorem is imported from the authors' prior work, and no ansatz is smuggled in via citation. The most plausible validity concern is the Section 6 augmentation of the corpus with 300 task-specific datasets for the semantic conditions, which could confound the quantitative success-rate comparison (10/12 vs 7/12 and 6/12). However, this is a methodological/statistical confound, not circularity: the reported successes are not equal to the injected corpus by construction, and the qualitative sensemaking findings are not derived from the augmentation. The paper's self-acknowledged limitations (small per-condition-order samples, restricted corpus, single keyword baseline) further indicate that the authors are not asserting a forced or self-validating result. Therefore the appropriate circularity finding is no significant circularity, score 0.

Assumptions & free parameters 4 free parameters · 4 assumptions · 0 invented entities

No new physical or mathematical entities are introduced. The only 'invented' items are system features (query reformulations, relevance indicators, concept filters), which are not entities in the sense of the ledger. Free parameters are implementation constants, not fitted to the study's outcome data; no derivations are involved.

free parameters (4)
  • Number of hypothetical schemas per query (k=3) = 3
    Hand-chosen in Section 5.2 to balance retrieval coverage and latency; not fitted to study outcomes.
  • k-means cluster count for reformulations and attribute concepts (k=15) = 15
    Hand-chosen clustering parameter in Sections 5.3.1 and 5.3.2; affects which suggestions are surfaced.
  • Top-k for relevance indicators (top-5) = 5
    Section 5.3.4 generates indicators for top-5 results with lazy evaluation for the rest; a design choice.
  • Number of attribute filter suggestions (5) and granularity filter suggestions (3) = 5 and 3
    Sections 5.3.2 and 5.3.3 surface fixed counts; design choices not fitted to data.
assumptions (4)
  • domain assumption Embedding cosine similarity is a valid proxy for semantic relevance between a query intent and a dataset
    The entire retrieval and filtering pipeline in Sections 5.2 and 5.3 assumes this.
  • domain assumption LLM-generated dataset metadata and relevance indicators are factually accurate enough for the study
    Sections 5.1 and 5.3.4 generate summaries, granularity tags, purposes, utilities and limitations without ground-truth validation.
  • domain assumption Participants' think-aloud comments and self-reported success reflect their genuine search experience
    Section 6 analysis relies on transcripts and quotes; standard for qualitative HCI studies.
  • domain assumption The 300-dataset augmentation (top-25 Kaggle results per task) does not invalidate task success comparisons
    Section 6 assumes this keeps the corpus relevant without biasing the interface comparison; this is the weakest of the study assumptions.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Rethinking Dataset Discovery with DataScout." pith.science (2026). https://pith.science/paper/R2B27J34

@misc{pith2026250718971,
  author       = {Pith},
  title        = {Pith review of: Rethinking Dataset Discovery with DataScout},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/R2B27J34}},
  note         = {Machine review of arXiv:2507.18971}
}
read the original abstract

Dataset Search -- the process of finding appropriate datasets for a given task -- remains a critical yet under-explored challenge in data science workflows. Assessing dataset suitability for a task (e.g., training a classification model) is a multi-pronged affair that involves understanding: data characteristics (e.g. granularity, attributes, size), semantics (e.g., data semantics, creation goals), and relevance to the task at hand. Present-day dataset search interfaces are restrictive -- users struggle to convey implicit preferences and lack visibility into the search space and result inclusion criteria -- making query iteration challenging. To bridge these gaps, we introduce DataScout to proactively steer users through the process of dataset discovery via -- (i) AI-assisted query reformulations informed by the underlying search space, (ii) semantic search and filtering based on dataset content, including attributes (columns) and granularity (rows), and (iii) dataset relevance indicators, generated dynamically based on the user-specified task. A within-subjects study with 12 participants comparing DataScout to keyword and semantic dataset search reveals that users uniquely employ DataScout's features not only for structured explorations, but also to glean feedback on their search queries and build conceptual models of the search space.

Figures

Figures reproduced from arXiv: 2507.18971 by the authors.

Figure 1
Figure 1. DataScout—a proactive dataset discovery interface. (A) Users begin by specifying their query as keywords, phrases, or complete sentences. (B) DataScout provides query reformulation suggestions to bridge the gap between the user’s query and datasets available in the search space. (C) Users may add exact matching-based or semantic filters, (D) search by attribute, apply (E) suggested attribute filters, or (F) suggeste… view at source ↗
Figure 2
Figure 2. Walkthrough of DataScout. Dana expresses her intent using the (A) getting started card. DataScout retrieves results. Dana reviews (B) query reformulation suggestions and hovers to view explanations. She clicks on the third suggestion— refreshing the results. Dana uses semantic (C) attribute and (D) granularity filter suggestions to narrow her search to datasets containing logged employee hours and country-level data… view at source ↗
Figure 3
Figure 3. Offline dataset collection, augmentation, embedding generation and indexing for [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: Online dataset search assistance. The user query is used to generate hypothetical schemas to retrieve matching [PITH_FULL_IMAGE:figures/full_fig_p008_4.png]
Figure 5
Figure 5. Figure 5: (A) Kaggle: Keyword dataset search condition [PITH_FULL_IMAGE:figures/full_fig_p015_5.png]
Figure 6
Figure 6. Figure 6: (B) Semantic Baseline: A stripped-down version of [PITH_FULL_IMAGE:figures/full_fig_p015_6.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

77 extracted references · 51 canonical work pages

  1. [1]

    Saleema Amershi, Dan Weld, Mihaela Vorvoreanu, Adam Fourney, Besmira Nushi, Penny Collisson, Jina Suh, Shamsi Iqbal, Paul N Bennett, Kori Inkpen, et al. 2019. Guidelines for human-AI interaction. In Proceedings of the 2019 chi conference on human factors in computing systems . 1–13

  2. [2]

    Marcia J Bates. 1989. The design of browsing and berrypicking techniques for the online search interface. Online review 13, 5 (1989), 407–424

  3. [3]

    Alex Bogatu, Norman W Paton, Mark Douthwaite, and André Freitas. 2022. Voyager: Data discovery and integration for data science. In Proceedings 25th International Conference on Extending Database Technology (EDBT 2022)

  4. [4]

    Virginia Braun and Victoria Clarke. 2006. Using thematic analysis in psychology. Qualitative research in psychology 3, 2 (2006), 77–101

  5. [5]

    Virginia Braun and Victoria Clarke. 2019. Reflecting on reflexive thematic analysis. Qualitative research in sport, exercise and health 11, 4 (2019), 589–597

  6. [6]

    Dan Brickley, Matthew Burgess, and Natasha Noy. 2019. Google Dataset Search: Building a search engine for datasets in an open Web ecosystem. In The world wide web conference. 1365–1375

  7. [7]

    Sonia Castelo, Rémi Rampin, Aécio Santos, Aline Bessa, Fernando Chirigati, and Juliana Freire. 2021. Auctus: A dataset search engine for data augmentation. arXiv preprint arXiv:2102.05716 (2021)

  8. [8]

    Adriane Chapman, Elena Simperl, Laura Koesten, George Konstantinidis, Luis- Daniel Ibáñez, Emilia Kacprzak, and Paul Groth. 2020. Dataset search: a survey. The VLDB Journal 29, 1 (2020), 251–272

Show all 77 references
  1. [9]

    Bhavya Chopra, Anna Fariha, Sumit Gulwani, Austin Z Henley, Daniel Perel- man, Mohammad Raza, Sherry Shi, Danny Simmons, and Ashish Tiwari. 2023. Cowrangler: Recommender system for data-wrangling scripts. In Companion of the 2023 International Conference on Management of Data ...

  2. [10]

    Mahdi Esmailoghli, Christoph Schnell, Renée J Miller, and Ziawasch Abedjan

  3. [11]

    Grace Fan, Jin Wang, Yuliang Li, and Renée J Miller. 2023. Table discovery in data lakes: State-of-the-art and future directions. In Companion of the 2023 International Conference on Management of Data . 69–75

  4. [12]

    Grace Fan, Jin Wang, Yuliang Li, Dan Zhang, and Renée Miller. 2022. Semantics- aware dataset discovery from data lakes with contextualized column-based representation learning. arXiv preprint arXiv:2210.01922 (2022)

  5. [13]

    Raul Castro Fernandez, Ziawasch Abedjan, Famien Koko, Gina Yuan, Samuel Madden, and Michael Stonebraker. 2018. Aurum: A data discovery system. In2018 IEEE 34th International Conference on Data Engineering (ICDE) . IEEE, 1001–1012

  6. [14]

    Sainyam Galhotra, Yue Gong, and Raul Castro Fernandez. 2023. Metam: Goal- oriented data discovery. In 2023 IEEE 39th International Conference on Data Engineering (ICDE). IEEE, 2780–2793

  7. [15]

    Saul Greenberg and Bill Buxton. 2008. Usability evaluation considered harmful (some of the time). In Proceedings of the SIGCHI conference on Human factors in computing systems. 111–120

  8. [16]

    Alex Groce, Todd Kulesza, Chaoqiang Zhang, Shalini Shamasunder, Margaret Burnett, Weng-Keen Wong, Simone Stumpf, Shubhomoy Das, Amber Shinsel, Forrest Bice, et al. 2013. You are the only possible oracle: Effective test selection for end users of interactive machine learning sy...

  9. [17]

    Philip J Guo, Sean Kandel, Joseph M Hellerstein, and Jeffrey Heer. 2011. Proac- tive wrangling: Mixed-initiative end-user programming of data transformation scripts. In Proceedings of the 24th annual ACM symposium on User interface software and technology. 65–74

  10. [18]

    Marti Hearst. 2009. Search user interfaces. Cambridge university press

  11. [19]

    Marti A Hearst. 2006. Clustering versus faceted categories for information exploration. Commun. ACM 49, 4 (2006), 59–61

  12. [20]

    Jonathan Herzig, Thomas Müller, Syrine Krichene, and Julian Martin Eisenschlos

  13. [21]

    Zezhou Huang, Jiaxiang Liu, Haonan Wang, and Eugene Wu. 2023. The Fast and the Private: Task-based Dataset Search. arXiv preprint arXiv:2308.05637 (2023)

  14. [22]

    Madelon Hulsebos, Wenjing Lin, Shreya Shankar, and Aditya Parameswaran

  15. [23]

    Sean Kandel, Andreas Paepcke, Joseph Hellerstein, and Jeffrey Heer. 2011. Wran- gler: Interactive visual specification of data transformation scripts. InProceedings of the sigchi conference on human factors in computing systems . 3363–3372

  16. [24]

    Harmanpreet Kaur, Eytan Adar, Eric Gilbert, and Cliff Lampe. 2022. Sensible AI: Re-imagining interpretability and explainability using sensemaking theory. In Proceedings of the 2022 ACM Conference on Fairness, Accountability, and Trans- parency. 702–714

  17. [25]

    Moe Kayali, Fabian Wenz, Nesime Tatbul, and Çağatay Demiralp. 2024. Mind the Data Gap: Bridging LLMs to Enterprise Data Integration. arXiv preprint arXiv:2412.20331 (2024)

  18. [26]

    Aamod Khatiwada, Grace Fan, Roee Shraga, Zixuan Chen, Wolfgang Gatterbauer, Renée J Miller, and Mirek Riedewald. 2023. Santos: Relationship-based semantic table union search. Proceedings of the ACM on Management of Data 1, 1 (2023), 1–25

  19. [27]

    Laura Koesten, Kathleen Gregory, Paul Groth, and Elena Simperl. 2021. Talking datasets–understanding data sensemaking behaviours. International journal of human-computer studies 146 (2021), 102562

  20. [28]

    Laura Koesten, Elena Simperl, Tom Blount, Emilia Kacprzak, and Jeni Tennison

  21. [29]

    Laura M Koesten, Emilia Kacprzak, Jenifer FA Tennison, and Elena Simperl

  22. [30]

    Andrew Kuznetsov, Joseph Chee Chang, Nathan Hahn, Napol Rachatasumrit, Bradley Breneisen, Julina Coupland, and Aniket Kittur. 2022. Fuse: In-situ sensemaking support in the browser. In Proceedings of the 35th Annual ACM Symposium on User Interface Software and Technology . 1–15

  23. [31]

    Andrew Kuznetsov, Michael Xieyang Liu, and Aniket Kittur. 2024. Tasks, Time, and Tools: Quantifying Online Sensemaking Efforts Through a Survey-based Study. arXiv preprint arXiv:2411.07206 (2024)

  24. [32]

    Bongshin Lee, Mary Czerwinski, George Robertson, and Benjamin B Bederson

  25. [33]

    Bongshin Lee, Greg Smith, George G Robertson, Mary Czerwinski, and Desney S Tan. 2009. FacetLens: exposing trends and relationships to support sensemaking within faceted datasets. InProceedings of the SIGCHI Conference on Human Factors in Computing Systems. 1293–1302

  26. [34]

    Aristotelis Leventidis, Martin Pekár Christensen, Matteo Lissandrini, Laura Di Rocco, Katja Hose, and Renée J Miller. 2024. A Large Scale Test Corpus for Semantic Table Search. In Proceedings of the 47th International ACM SIGIR Conference on Research and Development in Informa...

  27. [35]

    Michael Xieyang Liu, Tongshuang Wu, Tianying Chen, Franklin Mingzhe Li, Aniket Kittur, and Brad A Myers. 2024. Selenite: Scaffolding Online Sensemak- ing with Comprehensive Overviews Elicited from Large Language Models. In Proceedings of the 2024 CHI Conference on Human Factor...

  28. [36]

    Yu A Malkov and Dmitry A Yashunin. 2018. Efficient and robust approximate nearest neighbor search using hierarchical navigable small world graphs. IEEE transactions on pattern analysis and machine intelligence 42, 4 (2018), 824–836

  29. [37]

    Gary Marchionini. 2006. Exploratory search: from finding to understanding. Commun. ACM 49, 4 (2006), 41–46

  30. [38]

    Fengran Mo, Kelong Mao, Ziliang Zhao, Hongjin Qian, Haonan Chen, Yiruo Cheng, Xiaoxi Li, Yutao Zhu, Zhicheng Dou, and Jian-Yun Nie. 2024. A survey of conversational search. arXiv preprint arXiv:2410.15576 (2024)

  31. [39]

    Meredith Ringel Morris, Jarrod Lombardo, and Daniel Wigdor. 2010. WeSearch: supporting collaborative search and sensemaking on a tabletop display. In Pro- ceedings of the 2010 ACM conference on Computer supported cooperative work . 401–410

  32. [40]

    Donald A. Norman. 2002. The Design of Everyday Things. Basic Books, Inc., USA. Lin and Chopra et al

  33. [41]

    Paul Ouellette, Aidan Sciortino, Fatemeh Nargesian, Bahar Ghadiri Bashardoost, Erkang Zhu, Ken Q Pu, and Renée J Miller. 2021. RONIN: data lake exploration. Proceedings of the VLDB Endowment 14, 12 (2021)

  34. [42]

    Active Search

    Srishti Palani, Zijian Ding, Stephen MacNeil, and Steven P Dow. 2021. The" Active Search" Hypothesis: How search strategies relate to creative learning. In Proceedings of the 2021 conference on human information interaction and retrieval . 325–329

  35. [43]

    Srishti Palani, Zijian Ding, Austin Nguyen, Andrew Chuang, Stephen MacNeil, and Steven P Dow. 2021. CoNotate: Suggesting queries based on notes promotes knowledge discovery. InProceedings of the 2021 CHI Conference on Human Factors in Computing Systems. 1–14

  36. [44]

    Peter Pirolli and Stuart Card. 1999. Information foraging. Psychological review 106, 4 (1999), 643

  37. [45]

    Peter L. T. Pirolli. 2007. Information Foraging Theory: Adaptive Interaction with Information (1 ed.). Oxford University Press, Inc., USA

  38. [46]

    Filip Radlinski and Nick Craswell. 2017. A theoretical framework for conversa- tional search. In Proceedings of the 2017 conference on conference human informa- tion interaction and retrieval . 117–126

  39. [47]

    Corbin Rosset, Chenyan Xiong, Xia Song, Daniel Campos, Nick Craswell, Saurabh Tiwary, and Paul Bennett. 2020. Leading conversational search by suggesting useful questions. In Proceedings of the web conference 2020 . 1160–1170

  40. [48]

    Francesca Rossi, Kristen Brent Venable, and Toby Walsh. 2008. Preferences in constraint satisfaction and optimization. AI magazine 29, 4 (2008), 58–58

  41. [49]

    Daniel M Russell, Mark J Stefik, Peter Pirolli, and Stuart K Card. 1993. The cost structure of sensemaking. In Proceedings of the INTERACT’93 and CHI’93 conference on Human factors in computing systems . 269–276

  42. [50]

    Tony Russell-Rose and Tyler Tate. 2013. Chapter 2 - Information Seeking. In Designing the Search Experience, Tony Russell-Rose and Tyler Tate (Eds.). Morgan Kaufmann, 23–45. doi:10.1016/B978-0-12-396981-1.00002-1

  43. [51]

    Vidya Setlur, Andriy Kanyuka, and Arjun Srinivasan. 2023. Olio: A semantic search interface for data repositories. In Proceedings of the 36th Annual ACM Symposium on User Interface Software and Technology . 1–16

  44. [52]

    Ben Shneiderman. 1987. Designing The user interface: Strategies for effective human-computer interaction, 4/e (New Edition) . Pearson Education India

  45. [53]

    skrub data. 2025. skrub: A library for data cleaning and preprocessing. https: //github.com/skrub-data/skrub. Accessed: 2025-04-07

  46. [54]

    Greg Smith, Mary Czerwinski, Brian Meyers, Daniel Robbins, George Robertson, and Desney S Tan. 2006. FacetMap: A scalable search and browse visualization. IEEE Transactions on visualization and computer graphics 12, 5 (2006), 797–804

  47. [55]

    Katrina Sostek, Daniel M Russell, Nitesh Goyal, Tarfah Alrashed, Stella Dugall, and Natasha Noy. 2024. Discovering datasets on the web scale: Challenges and recommendations for Google Dataset Search. Harvard Data Science Review Special Issue 4 (2024)

  48. [56]

    Sangho Suh, Meng Chen, Bryan Min, Toby Jia-Jun Li, and Haijun Xia. 2024. Lumi- nate: Structured generation and exploration of design space with large language models for human-ai co-creation. In Proceedings of the 2024 CHI Conference on Human Factors in Computing Systems . 1–26

  49. [57]

    Sangho Suh, Bryan Min, Srishti Palani, and Haijun Xia. 2023. Sensecape: En- abling multilevel exploration and sensemaking with large language models. In Proceedings of the 36th annual ACM symposium on user interface software and technology. 1–18

  50. [58]

    Nitya Tarakad. 2024. A Peek Inside: How Snowflake’s New Universal Search Feature Was Built.Snowflake Builders Blog: Data Engineers, App Developers, AI/ML, & Data Science (February 2024). https://medium.com/snowflake/a-peek-inside- how-snowflakes-new-universal-search-feature-wa...

  51. [59]

    Daniel Tunkelang. 2022. Faceted search. Springer Nature

  52. [60]

    Iris Vessey. 1991. Cognitive fit: A theory-based analysis of the graphs versus tables literature. Decision sciences 22, 2 (1991), 219–240

  53. [61]

    Alexandra Vtyurina, Denis Savenkov, Eugene Agichtein, and Charles LA Clarke

  54. [62]

    Chi Zhang. 2024. Adding Intelligence to Databricks Search. Databricks Blog (March 2024). https://www.databricks.com/blog/adding-intelligence-to- databricks-search

  55. [63]

    Yongfeng Zhang, Xu Chen, Qingyao Ai, Liu Yang, and W Bruce Croft. 2018. To- wards conversational search and recommendation: System ask, user respond. In Proceedings of the 27th acm international conference on information and knowledge management. 177–186

  56. [64]

    Yihang Zhao, Albert Meroño-Peñuela, and Elena Simperl. 2024. User Experience in Dataset Search Platform Interfaces. arXiv e-prints (2024), arXiv–2403. Rethinking Dataset Discovery with DataScout A Screenshots of User Study Conditions Here, we present screenshots for our baseli...

  57. [68]

    In Proceedings of the 2017 chi conference extended abstracts on human factors in computing systems

    Exploring conversational search with humans, assistants, and wizards. In Proceedings of the 2017 chi conference extended abstracts on human factors in computing systems. 2187–2193

  58. [72]

    Description Summary: In 1–3 sentences, provide a brief and summarized description of the dataset

  59. [73]

    training a regression model

    Purposes: Provide a list of analytical, data science, visual- ization, or machine learning tasks that can be performed with this dataset. e.g., ["training a regression model", "temporal analysis"]

  60. [74]

    If no information is available about the source of the data, output ‘N/A’

    Dataset Source & Collection Methods: Gather the source(s) of this dataset, which could include names and/or affiliations of persons, website URLs, web-APIs, synthetic sources, human annotations, and so on. If no information is available about the source of the data, output ‘N/A’

  61. [75]

    description_summary

    Column Descriptions: For each column in the dataset, pro- vide a brief description for the column with its data type. Output Schema: {“description_summary”: string, “dataset_purposes”: list[string], “dataset_sources”: string, “column_descriptions”: list[{“column_name”: string,...

  62. [76]

    No significant utilities

    Utilities: Identify the strongest factors that make this dataset useful. Look for the presence of relevant attributes, high data quality, and matching intent. If there are no strong advantages, return "No significant utilities."

  63. [77]

    dataset only con- tains records of location X

    Limitations: Identify limitations such as missing relevant at- tributes, specific geographical locations (e.g., “dataset only con- tains records of location X”), specific temporal ranges (e.g., “data belongs to X and Y time range”), poor data quality and missing or incomplete ...

  64. [2005]

    In CHI’05 extended abstracts on Human factors in computing systems

    Understanding research trends in conferences using PaperLens. In CHI’05 extended abstracts on Human factors in computing systems . 1969–1972

  65. [2017]

    In Proceedings of the 2017 CHI conference on human factors in computing systems

    The Trials and Tribulations of Working with Structured Data: -a Study on Information Seeking Behaviour. In Proceedings of the 2017 CHI conference on human factors in computing systems . 1277–1289

  66. [2020]

    International journal of human-computer studies 135 (2020), 102367

    Everything you always wanted to know about a dataset: Studies in data summarisation. International journal of human-computer studies 135 (2020), 102367

  67. [2021]

    arXiv preprint arXiv:2103.12011 (2021)

    Open domain question answering over tables via dense retrieval. arXiv preprint arXiv:2103.12011 (2021)

  68. [2023]

    arXiv preprint arXiv:2310.02656 (2023)

    Blend: A unified data discovery system. arXiv preprint arXiv:2310.02656 (2023)

  69. [2024]

    In Proceedings of the 2024 Workshop on Human-In-the-Loop Data Analytics

    It took longer than I was expecting: Why is dataset search still so hard?. In Proceedings of the 2024 Workshop on Human-In-the-Loop Data Analytics . 1–4

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.