Michael Markert

25 Followers
20 Following
55 Posts

Ask your data any question - on your own computer
I made a tutorial on how to use local LLMs in #openrefine via the API of LM Studio (but Anything LLM, Jan, ... will do as well).
It is in German but has English subtitles.

https://m.youtube.com/watch?v=jQLcthiOli8

Tutorial zur Nutzung eines lokalen LLM in OpenRefine für NER, Bereinigung, Anreicherung

YouTube
Since July I contribute to @openstreetmap using @streetcomplete. It is a fantastic way to learn about the #opendatacommunity #mapping #urbanaccessibility great #ux and your surroundings, naturally. Today I did edit no. 10.000 in #erfurt Ilversgehofen.
#openstreetmap #streetcomplete
Interested in former territories a place was part of, e.g. for #provenanceresearch? Here is a #JupyterNotebook to query for is_in relations in #openhistoricalmap and download the results including links to #wikidata and #wikipedia as a CSV file. https://github.com/MichaelMarkert/GND4C/tree/main/historicalPlaces
@ohm
#provenienzforschung #datawrangling #linkedopendata #APIsyoumightlove
GND4C/historicalPlaces at main · MichaelMarkert/GND4C

Contribute to MichaelMarkert/GND4C development by creating an account on GitHub.

GitHub
The #wikidata UI is focused on editing not reading so I wrote some custom #CSS to use with the #Stylebot browser add-on. This way you get a less cluttered, compact layout and even can highlight your favourite properties.
Find the demo CSS here: https://github.com/MichaelMarkert/GND4C/tree/main/Wikidata_CSS
@wikidata
GND4C/Wikidata_CSS at main · MichaelMarkert/GND4C

Contribute to MichaelMarkert/GND4C development by creating an account on GitHub.

GitHub
Zur einfacheren Recherche von Provenienzmerkmalen in der #gnd im Rahmen der #ThULB #provenienzforschung habe ich für die Kolleg:innen eine kleine #Streamlit -App geskriptet, die via #lobid die GND durchsucht, aus den Datensätzen die Quellangaben abfragt und die zugehörigen Bilder aus dem #GBV #Provenienz Wiki und aus der Deutschen Fotothek der #slubdresden harvestet.
Falls das sonst noch jemand nützlich findet, hier die Datei: https://github.com/MichaelMarkert/GND4C/tree/main/Provenienz
@lobid @SLUBDresden
GND4C/Provenienz at main · MichaelMarkert/GND4C

Contribute to MichaelMarkert/GND4C development by creating an account on GitHub.

GitHub
Schon mal daran gedacht, mit #gpt direkt eigene Daten zu befragen? In #openrefine etwa? Ich habe es heute einmal probiert und die Ergebnisse sind sehr vielversprechend - zuverlässige #ner praktisch ohne Code z. B. Hier meine Anleitung: https://forum.openrefine.org/t/using-the-openai-api-to-apply-natural-language-queries-to-cells-data/308
Using the OpenAI API to apply natural language queries to cells/data

I am fiddling around with the OpenAI API in Python for named entity recognition and just had the idea to use it directly within OpenRefine. You just need to create an account at http://openai.com, generate an API key, and copy it to “YOUR-API-KEY” in the code below. Now you can formulate a question on your data in the “prompt” and apply it to the values in the respective column. You even can define a format for your response in the query, here it is JSON(ish). The code, by the way, for the mo...

OpenRefine