Exploratory Unstructured Data Analysis: A Formative Study and Implications for Human-AI Collaboration | TU Wien

Information

Publication Type: Conference Paper
Workgroup(s)/Project(s):
- Joint Human-Machine Data Exploration
Date: October 2026
ISBN: 979-8-4007-2373-5
Publisher: Association for Computing Machinery
Open Access: yes
Location: Vaasa
Lecturer: Johannes Eschner
Event: 14th Nordic Conference on Human-Computer Interaction (NordiCHI 2026)
DOI: 10.1145/3829807.3829882
Booktitle: NordiCHI '26 : Pulse - exploring the energy that drives us : Proceedings of the 14th Nordic Conference on Human-Computer Interaction
Pages: 16
Conference date: 3. October 2026 – 7. October 2026
Keywords: Exploratory data analysis, Visual Analytics, mixed-methods evaluation

Abstract

We propose a conceptual framework for exploratory data analysis of (large) unstructured data (EluDA), combining classical elements (querying, visualization) with active knowledge construction in the “search for structure”. In a formative study, users conceptualized a structure for an image dataset during exploration. We found that users conceptualize by building faceted classifications bottom-up and rarely create meaningful spatial categorization during this process. We also evaluated CLIP for zero-shot assignment and semantic categorization, finding that it remains unreliable for assigning user-defined concepts to images but does support semantic grouping. Based on these findings, we identify and discuss four key opportunities for human-AI collaboration in EluDA: intelligent sampling and visualization to maximize data visibility; incremental and few-shot learning to minimize effort for reliable assignment; automatic category, concept, and facet suggestions to reduce effort during the search for structure; and the necessity for effective trust calibration methods.

Additional Files and Images

paper

Weblinks

BibTeX

@inproceedings{eschner-2026-eud,
  title =      "Exploratory Unstructured Data Analysis: A Formative Study
               and Implications for Human-AI Collaboration",
  author =     "Johannes Eschner and Dominik Eitler and Max Irendorfer and
               Patrick Kramml and Matthias Zeppelzauer and Manuela Waldner",
  year =       "2026",
  abstract =   "We propose a conceptual framework for exploratory data
               analysis of (large) unstructured data (EluDA), combining
               classical elements (querying, visualization) with active
               knowledge construction in the “search for structure”. In
               a formative study, users conceptualized a structure for an
               image dataset during exploration. We found that users
               conceptualize by building faceted classifications bottom-up
               and rarely create meaningful spatial categorization during
               this process. We also evaluated CLIP for zero-shot
               assignment and semantic categorization, finding that it
               remains unreliable for assigning user-defined concepts to
               images but does support semantic grouping. Based on these
               findings, we identify and discuss four key opportunities for
               human-AI collaboration in EluDA: intelligent sampling and
               visualization to maximize data visibility; incremental and
               few-shot learning to minimize effort for reliable
               assignment; automatic category, concept, and facet
               suggestions to reduce effort during the search for
               structure; and the necessity for effective trust calibration
               methods.",
  month =      oct,
  isbn =       "979-8-4007-2373-5",
  publisher =  "Association for Computing Machinery",
  location =   "Vaasa",
  event =      "14th Nordic Conference on Human-Computer Interaction
               (NordiCHI 2026)",
  doi =        "10.1145/3829807.3829882",
  booktitle =  "NordiCHI '26 : Pulse - exploring the energy that drives us :
               Proceedings of the 14th Nordic Conference on Human-Computer
               Interaction",
  pages =      "16",
  keywords =   "Exploratory data analysis, Visual Analytics, mixed-methods
               evaluation",
  URL =        "https://www.cg.tuwien.ac.at/research/publications/2026/eschner-2026-eud/",
}