Digital Democracy Lab (@digdemlab) 's Twitter Profile
Digital Democracy Lab

@digdemlab

We research digital technology and democracy •• PIs: @fgilardi @LucasLeemann @ARGohdes @karstendonnay @ellliottt •• Developer: @MaelKubli •• Part of @UZH_dsi

ID: 951765794156957696

linkhttps://digdemlab.io/ calendar_today12-01-2018 10:40:23

5,5K Tweet

3,3K Followers

2,2K Following

Oriol J. Bosch (@orioljbosch) 's Twitter Profile Photo

🚨 Do you use digital trace data / web tracking data? Do you want to measure what people do online? We published a Total Error Framework for web tracking data: bit.ly/3hj0qTB We show: how to use it, errors to expect & best practices informed by a real case study! 🧵

🚨 Do you use digital trace data / web tracking data? Do you want to measure what people do online? 

We published a Total Error Framework for web tracking data: bit.ly/3hj0qTB

We show: how to use it, errors to expect & best practices informed by a real case study!
🧵
Aleksandra Urman (@aurman21) 's Twitter Profile Photo

Come be my colleague! Our Social Computing Group (University of Zurich) led by Ancsa Hannak has a Postdoc position open. If you are looking for a postdoc position in Computational Social Science or Social Computing, apply! And reach out if you have questions! jobs.uzh.ch/offene-stellen…

Mauricio Mandujano Manriquez (@mauriciomandu7) 's Twitter Profile Photo

On the last week of October I had the privilege of attending the Digital Democracy Workshop 2022 by Digital Democracy Lab IPZ - Political Science UZH. There I presented on working paper on the Compatibility of the use of Machine Learning Models in democratic societies and the implications for political theory.

On the last week of October I had the privilege of attending the Digital Democracy Workshop 2022 by <a href="/DigDemLab/">Digital Democracy Lab</a> <a href="/IPZ_ch/">IPZ - Political Science UZH</a>. There I presented on working paper on the Compatibility of the use of Machine Learning Models in democratic societies and the implications for political theory.
Fabrizio Gilardi (@fgilardi) 's Twitter Profile Photo

Monetization of social media engagements increases the sharing of false (and other) news but penalization moderates it New PRODIGI paper, together with Meysam Alizadeh (lead author) and Emma Hoes osf.io/har52/

Monetization of social media engagements increases the sharing of false (and other) news but penalization moderates it

New <a href="/PRODIGI_ERC/">PRODIGI</a> paper, together with <a href="/MeysamAIizadeh/">Meysam Alizadeh</a> (lead author) and <a href="/EmmaHoes93/">Emma Hoes</a>

osf.io/har52/
Silja Häusermann (@siljahausermann) 's Twitter Profile Photo

For the area of "digitalization", Emma Hoes brilliantly presented her research on the responsiveness of social media platforms to media reporting on them, as well as on the effects on user behavior of monetization of social media content. IPZ - Political Science UZH

For the area of "digitalization", <a href="/EmmaHoes93/">Emma Hoes</a> brilliantly presented her research on the responsiveness of social media platforms to media reporting on them, as well as on the effects on user behavior of monetization of social media content. <a href="/IPZ_ch/">IPZ - Political Science UZH</a>
Andrea Ceron (@andreaceron83) 's Twitter Profile Photo

The #Encyclopedia of #Technology and #Politics is finally out! 94 great scholars from 5 continents (47% female; 46% junior) talk about lights and shades of digital & tech revolution e-elgar.com/shop/gbp/elgar… 🧵

Fabrizio Gilardi (@fgilardi) 's Twitter Profile Photo

Human-in-the-Loop Hate Speech Classification in a Multilingual Context New paper with Ana Kotarcic, Dominik Hangartner, Selina Kurer & Karsten Donnay Ana Kotarcic is presenting it at #EMNLP2022 this week, check it out if you're attending arxiv.org/abs/2212.02108

Human-in-the-Loop Hate Speech Classification in a Multilingual Context

New paper with Ana Kotarcic, Dominik Hangartner, Selina Kurer &amp; <a href="/karstendonnay/">Karsten Donnay</a> 

Ana Kotarcic is presenting it at #EMNLP2022 this week, check it out if you're attending

arxiv.org/abs/2212.02108
Fabrizio Gilardi (@fgilardi) 's Twitter Profile Photo

ChatGPT Outperforms Crowd-Workers for Text-Annotation Tasks with Meysam Alizadeh Maël Kubli We find that zero-shot #ChatGPT: > has better accuracy than #MTurk > has better intercoder agreement than MTurk and trained coders > is 20x cheaper than MTurk arxiv.org/abs/2303.15056

ChatGPT Outperforms Crowd-Workers for Text-Annotation Tasks

with <a href="/MeysamAIizadeh/">Meysam Alizadeh</a> <a href="/MaelKubli/">Maël Kubli</a> 

We find that zero-shot #ChatGPT:
&gt; has better accuracy than #MTurk
&gt; has better intercoder agreement than MTurk and trained coders
&gt; is 20x cheaper than MTurk

arxiv.org/abs/2303.15056
Fabrizio Gilardi (@fgilardi) 's Twitter Profile Photo

ChatGPT Can Replace the Underpaid Workers Who Train AI, Researchers Say VICE article on our work on #ChatGPT annotation capabilities Meysam Alizadeh Maël Kubli vice.com/en/article/ak3…

DSI Digital Democracy (@dsi_democracy) 's Twitter Profile Photo

⚡️Call for papers⚡️ Digital Democracy Workshop (4th edition) October 26-27, 2023 University of Zurich DSI Digital Democracy Digital Democracy Lab ⏰ Application deadline: May 31, 2023 🙌 Please share widely! digdemlab.io/forms/workshop…

⚡️Call for papers⚡️

Digital Democracy Workshop (4th edition)
October 26-27, 2023
University of Zurich

<a href="/DSI_Democracy/">DSI Digital Democracy</a>
<a href="/DigDemLab/">Digital Democracy Lab</a>
⏰ Application deadline: May 31, 2023
🙌 Please share widely!

digdemlab.io/forms/workshop…
Fabrizio Gilardi (@fgilardi) 's Twitter Profile Photo

⚡️ New paper ⚡️ "Open-source LLMs such as HuggingChat and FLAN exhibit text-annotation performance metrics that generally exceed those of MTurk and rival those of ChatGPT" with meysam alizadeh @maelkubli et al. arxiv.org/abs/2307.02179 #ChatGPT #LLMs #OpenSource #NLP

⚡️ New paper ⚡️

"Open-source LLMs such as HuggingChat and FLAN exhibit text-annotation performance metrics that generally exceed those of MTurk and rival those of ChatGPT"

with <a href="/meysamalizadeh/">meysam alizadeh</a> @maelkubli et al.

arxiv.org/abs/2307.02179

#ChatGPT #LLMs #OpenSource #NLP
Fabrizio Gilardi (@fgilardi) 's Twitter Profile Photo

Very happy that our paper "ChatGPT outperforms crowd workers for text-annotation tasks" is now out at PNASNews 😀 with Meysam Alizadeh & Maël Kubli pnas.org/doi/10.1073/pn… #ChatGPT #LLMs

Very happy that our paper "ChatGPT outperforms crowd workers for text-annotation tasks" is now out at <a href="/PNASNews/">PNASNews</a> 😀

with <a href="/MeysamAIizadeh/">Meysam Alizadeh</a> &amp; <a href="/MaelKubli/">Maël Kubli</a>

pnas.org/doi/10.1073/pn… 

#ChatGPT #LLMs
Fabrizio Gilardi (@fgilardi) 's Twitter Profile Photo

⚡️Job alert ⚡️ 2-year postdoc position, part of my PRODIGI project IPZ - Political Science UZH > Digital tech & politics > Causal inference or computational social science > PhD completed or close to completion > Starting Jan 2024 (or later) Happy to answer any Qs! fabriziogilardi.org/resources/jobs…

⚡️Job alert ⚡️

2-year postdoc position, part of my <a href="/PRODIGI_ERC/">PRODIGI</a>
project <a href="/IPZ_ch/">IPZ - Political Science UZH</a>

&gt; Digital tech &amp; politics
&gt; Causal inference or computational social science
&gt; PhD completed or close to completion
&gt; Starting Jan 2024 (or later)

 Happy to answer any Qs!

fabriziogilardi.org/resources/jobs…