EDITED v. 46
To Be Renamed!
Identifying and Improving Generic or Incorrect Object Names in Museum Databases
| 8 | 59 | 49 |
|---|
Event finish
Research
Reconciling the "Gebrauch" (use) column of the data with Wikidata worked well when the values followed the usual pattern.
It was more difficult when the "use" didn't make sense without the object title to guide us.

While testing methods, we also asked ChatGPT for classifications using the prompt: "Attached are descriptions from a German ethnographic collection of over 60,000 objects. Assist in generating categorizations by clustering the entries based on its internal contents, and also compare with the categorizations other German/similar ethnographic collections".

Created an text visualization for both the Objekt Name and the Karte Gebrauch columns to see the most common labels from the original database.

JOINED
Most of the time it does work and saves us a lot of time with cleaning and labeling the data.

Sometimes auto-clustering can go wrong...
Ideas for further exploration
Two possible approaches:
-
Create a new field for subject headings → generic term Investigate whether there are established vocabularies that could be adopted (or adapted). Review how other museums handle subject headings and terminologies.
-
Use the “Object Name” field → specific term (as specific as possible) Review concrete examples from Tim’s comparative cases to determine how specific object designations are applied in practice.

The VMZ's catalogue database can export the images of a subset of items (the result of a query, like "All items with question marks in their title"). Unfortunately, it can only export the images as a PDF. I wrote a quick script to extract all the images from the PDF and save them with the item id as the file name: https://gist.github.com/bellisk/c0d80fda6d21d14d45a37fffa4f1a333
We experimented with OpenRefine to rename objects (see screnshot). We also experimented with the application Panoptic (by CERES) for image similarity in locally stored pictures (see screenshot).



break-through: Tim used embedded models and public API's to grab object photos from other collections - allowed to do a visual comparison of all VMZ-objects which contain a "?" in the object name
for IIIF newbies: https://medium.com/@adam.coulson/first-steps-into-iiif-8eb5f33b2b18
new approach!!: is there a IIIF Manifest for image similarity?

... and: Also no keywording/tagging function in the current database, limiting systematic indexing, searching, and terminology management.
Main challenge: The revision of generic terms is hampered by the absence of both a controlled vocabulary and a thesaurus
another idea: Systematically review objects with temporary “o.Nr.” designations and match them against objects currently recorded as missing, with the aim of re-identifying and assigning proper inventory numbers.
(o.Nr. means: "object without a [proper] object number", e.g. objects with temporary numbers pending identification)
Put together a small set of images to test reverse image research - including the "Schnabelgeiss"
Working in OpenRefine on the quotation marks:
- added new colum in excel
- replace (>> Find / >> Replace) all kinds of quotation mark with straigh ones: "...."
JOINED
This morning we decided to start working on two lines:
- reverse image research for all objects that contain a "?" in the field
- Using OpenRefine: work on (cluster / clean) object names that contain 1. question marks, 2. quotation marks (in all versions: German/English/Guillement/single-double etc.)
Happy to have on board with us!: Christa Ackermann Pierre-Louis Blanchard Rae Knowler Kapitolina Kostina Tim Ruediger Nole Wade
Project
JOINED
Start
PDF eingefügt
bild gelöscht
bild eingefügt
GLAMhack 2026
The program we first tried on local storage was limited in capability. Then we did data cleaning on names, experimenting with OpenRefine (a very cool tool, we thought). A tool was vibe coded to take pictures from various collections online, compare images and their identifiers. Now we are going to work on how to prepare this data for curators, to research the regions and departments of the objects 🏺