Sal Hagen
banner
hetsalwel.bsky.social
Sal Hagen
@hetsalwel.bsky.social
230 followers 110 following 41 posts
Postdoctoral researcher @ UvA | @d-eepculture.bsky.social | @digitalmethods.net | @oilab.bsky.social
Posts Media Videos Starter Packs
When searching for thematic questions (e.g. 'who should I vote for if I want less taxation'), the D66 party website ranked as the top source, cited even more than Wikipedia.

For more information on the data collection, see our Deep Culture project website: deep-culture.org/google-ai-ov...
With @best.wel.moe and @nieuwsuur.nl, we studied Google's AI Overviews for political searches in the NL. We found that larger parties often benefit and that Google changed its use of AI Overviews *three times* during the run-up to the Dutch elections.

Read the article here: nos.nl/nieuwsuur/ar...
Google zet gekleurde AI-antwoorden midden in campagnetijd aan, uit, en weer aan
GroenLinks-PvdA, D66 en de VVD worden in de antwoorden op algemene politieke vragen het vaakst genoemd. Kleinere partijen komen veel minder vaak aan bod.
nos.nl
Reposted by Sal Hagen
Keywords are not neutral.

With Tobias Blanke, we investigate the power of keywords as interpretive tools in our work "Keywords in Digital Humanities: A critical assessment of computational techniques for mapping security and freedom in historical debates".

You can read the article now on DHQ!
DHQ: Digital Humanities Quarterly: Keywords in Digital Humanities – A critical assessment of computational techniques for mapping security and freedom in histor...
dhq.digitalhumanities.org
Reposted by Sal Hagen
Super fun to contribute to this brief interview with Terminally Online, a student-led organization on New Media at the University of Amsterdam.

We spoke about my wonderful team‘s research (@d-eepculture.bsky.social), who inspires me, and some of my own work.

🔗: www.linkedin.com/pulse/interv...
Interview with Louis Ravn
Louis Ravn is a PhD Candidate on Deep Culture at the University of Amsterdam, working in the fields of science and technology studies and critical data/AI studies. Moreover, he is assistant editor for...
www.linkedin.com
Reposted by Sal Hagen
Dit is de omgekeerde wereld. De academische vrijheid wordt bedreigd doordat onze onderzoekers die zich uitspreken in het publieke debat over “omstreden” onderwerpen als snel het doelwit worden van online bedreigingen, met name uit extreemrechtse hoek.
'Nederlandse universiteiten schuwen debat met andersdenkenden'
Is er op Nederlandse universiteiten nog wel ruimte voor een open debat? "Radicale vrijheid van meningsuiting bestaat in Nederland niet."
nos.nl
Reposted by Sal Hagen
🚨The Critical AI Seminar Series returns in 2025/2026 ✨

This year we will have four talks and are beyond excited to welcome @amoorelouise.bsky.social & @alexcampolo.bsky.social, Fabian Offert, Helene Ratner & @nannathylstrup.bsky.social and @thaophan.bsky.social

Sign up: bit.ly/uvacriticala...
Reposted by Sal Hagen
Together with Marc Tuters and @robtoes.bsky.social I wrote an essay on the Charlie Kirk assassination, the Groypers - Nick Fuentes connection and the state of memetic political violence today oilab.eu/frozen-antag...
Frozen Antagonisms: Charlie Kirk, Groypers, and the Meme Politics of Violence – OILab
oilab.eu
The last few weeks I worked 4CAT's LLM features. We're not free from Silicon Valley APIs yet, but you have more options in terms of providers (including Mistral) and it's also possible to use local LLMs through LM Studio and Ollama.

Check it out:
4cat.nl 4CAT @4cat.nl · Jul 18
🧠 New: 'LLM Prompter' processor
A much more robust and flexible way to analyse datasets with LLMs
• More APIs: OpenAI, Google, Anthropic, and Mistral
• Support for LM Studio and Ollama to use local (and free) LLMs
• Batching of input values to increase speed and reduce cost
• Structured outputs
This was also one of the more interesting methodologies I've worked on. We sourced ~3000 questions from 4chan and leftychan, rephrased them through LLMs, queried them on Google and Bing, and scraped the SERP contents through Selenium, 4CAT, and Zoekplaatje.

Code: github.com/sal-uva/radi...
Google relies heavily on Reddit, showing up in 75% of SERPs(!). Controversial topics that are not enriched are thus often "relayed" to the amateur interpretations of Redditors (see below).

We call for continued SERP audits to interrogate the volatility of enrichment amidst changing political tides.
We find that controversial searches get less components. Some keywords were even blacklisted from enrichment altogether, including "trump" and "biden".

This suggests Google and Bing tend avoid burning their fingers on certain political topics, raising concerns on the reliability of these systems.
🔎New paper! Guillén Torres and I study "enrichment" on Google and Bing as a form of content moderation.

Main finding: enrichment can be erratic, with components like "AI Overviews" even being banned for certain political queries like those including "trump".

Pre-print: ssrn.com/abstract=528...
We find that controversial searches get less components. Some keywords were even blacklisted from enrichment altogether, including "trump" and "biden".

This suggests Google and Bing tend avoid burning their fingers on certain political topics, raising concerns on the reliability of these systems.
Op nos.nl ging de laatste vijftien jaar even vaak over klimaatverandering als over Erik ten Hag.

(2853 vs. 2766 artikelen/updates)

@xrnl.bsky.social @nieuws.nos.nl
Tired of looking at Excel rows when investigating social media data? 4CAT's new Explorer feature is here to help!
4cat.nl 4CAT @4cat.nl · Jun 4
🔎The Explorer lets you examine datasets in a rich environment where social media posts are "reassembled" to look like their original appearance.

You can sort data, make manual annotations like field notes, or let (LLM-)processors add categories and notes.

How to use: github.com/digitalmetho...
yes een nieuwe kans voor Schier
Nog steeds aan het bijkomen van de race tussen Schier en Vlieland
Juist het minst door AD en de Telegraaf! De grafiek is 'stacked' dus hun aandeel is het minst tov NRC en de Volkskrant. Maar andere manieren van meten zouden interessant zijn inderdaad (bv. om te corrigeren voor lengte van artikelen of te kijken hoe centraal het thema is).
Nee, omdat je geen volledig archief tot je beschikking hebt. Maar wat wel mogelijk is is om andere trefwoorden pakken en het daarmee vergelijken (bv. "ajax").
tijd om NRC te shorten
Na ons artikel uit 2021 over klimaatverslaggeving (www.groene.nl/artikel/we-h...) was het tijd voor een data-update.

Gelukkig is meer aandacht voor de klimaatcrisis in de vier grootste kranten. Maar er is nog steeds een golfbeweging gaande en aandacht is de laatste maanden weer afgenomen.