NERDS

#kv25 Since I keep reading and reviewing papers using #LLMs to simulate social dynamics I think it's just fair to see how they would do in simulating the outcome of the local elections we had yesterday in #Copenhagen. The plot below shows the "predicted" value obtained via gpt4o

🔈 *Call for Participation* 🔈

Italian Conference on Computational Social Science – CS2Italy
May 19-21, 2026 in Torino, Italy

Premier venue for Computational Social Science in Italy: #sociology #polisci #economics #netsci #complexsystems ++

Submit your abstract by Jan 15: cs2italy.org

plz reshare

Yay, datasci.social is now 3 years old! 🥳
Same as last year, so I don't bother with a new post: https://community.datasci.social/blog/2024-11-15/two-years-old/
stable users - no issues - 99.9+% uptime - muchas gracias to our supporters!! 💖

Only new thing: Given bluesky's predictable moderation […]

A network graph

A short thread of observations using Max Andrew's tool to explore the network of relationships revealed in the Epstein estate email cache. @thatsjustlikeyouropinionman.com
🧵
1. There is a tight inner circle, densely connected. Trump is in the near outer rim.
epstein-doc-explorer-1.onrender.com

Picard management tip: As soon as you find out that a person is truly toxic, get them the hell off your ship.

I just got 3.1 mil DKK, so if you are interested in uncovering how climate change 🏭☁️⛅⛈️🌊🌪️🌬️ affects our movement patterns and mobility related behaviors, then I will be opening a PhD position later this year

Best poster I've ever seen -- check how the shape reflects the contents! (By Anastasia Vybornova & @trivikrama.bsky.social) #ODISSEI2025

A screenshot of Catherine Yeh's website: https://catherinesyeh.github.io/ The website looks clean, colorful, and modern.

A screenshot of four of the websites included in my Are.na board. They mostly show academic websites in different styles. Some are more text-heavy, some feature more colors and images.

If you're a student in need of a personal website (and if you're doing research, yes, you need a website!), I keep a list of nice examples here, most of which are reusable: www.are.na/maria-antoni...

For example, I just spotted this beautiful website by Catherine Yeh: github.com/catherinesye...

Portrait photo of a smiling researcher, grey background

Vedran wins Independent Research Fund Denmark Grant!
nerds.itu.dk/2025/11/03/v...

Small multiples plot of different street networks simplified with various algorithms

Left: Edge activations for the set of influencers identified by CHD. Edges are colored and sized according to how often they are activated during simulations. Nodes colored black are seed nodes. Right: Edge activations for one of the fairer seed sets identified by our algorithm.

Two new NERDS papers: Algorithmic bias, street network simplification
nerds.itu.dk/2025/11/03/t...

Detecting bias in algorithms used to disseminate information in social networks and mitigating it using multiobjective optimization Abstract. Social connections are conduits through which individuals communicate, information propagates, and diseases spread. Identifying individuals who a

How fair are the algorithms that decide who gets information online? Our new paper in PNAS Nexus shows that maximization algorithms, the ones used to pick “influencers” in social networks, unintentionally deepen inequality.

🧠 Sekara, Dotu, Cebrian, Moro & García-Herranz
🔗 doi.org/10.1093/pnas...

SODAS Data Discussion 3 (Fall 2025) SODAS is delighted to host Daniel Juhász Vigild and Stephanie Brandl for the Fall 2025 Data Discussion series!

Join us for a Data Discussion on Friday, November 7! 📅

Daniel Juhász Vigild will start by exploring how government use of AI impacts its trustworthiness, while Stephanie Brandl will examine whether LLMs can identify and classify fine-grained forms of populism.

Event🔗: sodas.ku.dk/events/sodas...

Find his lecture slides here: uscuni.org/talks/slides...

Martin gave a fantastic talk today about his personal path and insights about #opensource in front of a 100+ people audience, impressing also many students!

Martin Fleischmann @martinfleischmann.net will present in 1.5 hours at ITU, AUD 1: "Open source inside out"

Excited to share that my first journal paper (w/ Dr Craig Robson & @alistaircford.bsky.social) has been published in the Journal of Cycling and Micromobility Research! Read more (open access) here: doi.org/10.1016/j.jc...

Thanks to @geospatialcdt.bsky.social and Craig and Ali for enabling this!

Kilde: Mobilitetsredegørelsen 2025 På cykelområdet er der ca. blevet bevilliget 160 mio. kr. årligt (10-års gennemsnit for 2016-2025). Investeringer på cykelområdet omfatter både midler til cykelinfrastruktur og cykelinitiativer, men også sikre skoleveje, trafiksikkerhed samt eksterne investeringer og bidrag fra puljer. I 2025 blev der bevilliget 174 mio. kr. til cykelområdet. Af de 174 mio. kr. blev 82 mio. kr. bevilliget primær til cykelinfrastruktur og cykelinitiativer, 68 mio. kr. til sikre skoleveje, trafiksikkerhed, strøggader, cykel- og gangbroer og 11 mio. kr. i forbindelse med eksterne investeringer fra trafikselskaber samt bidrag fra puljer.

From an urban planning perspective it is very good and much needed that Copenhagen is increasing its bicycle budget. It is especially important that some of this investment will be allocated to child safety, as Copenhagen's transport policies have indeed neglected children. I personally experience this every morning through the terrifying traffic chaos on the street of my 7 year old daughter's school. As a consequence, Copenhagen has fallen behind several cities in Europe. Investing into cycling is increasing a city's livability, and also comes with massive economic benefits, especially to the public health system. We should however not forget, that still vehicular infrastructure receives a budget that is many times higher than cycling, so this bicycle budget is only like a drop in the bucket. Despite the laudable efforts, Copenhagen is still missing the political will to boldly transform its transport system towards becoming truly human-centric. Also the public wants this, as one can see clearly in Paris' ongoing success story. So, despite offering fantastic cycling infrastructure for fit adults, Copenhagen unfortunately remains inherently car-centric. A budget increase is great, but not enough by far - it will need a fundamental rethinking of transport, urban living, and crucially the political will to change this.

Here the city's press release: https://via.ritzau.dk/pressemeddelelse/14560655/line-barfod-om-kobenhavnsk-budget-stolt-af-det-storste-cykelbudget-nogensinde

They ~quadrupled the bicycle budget to >600 mio. DKK. Below my full comments to the journalist, which was used for the article.

Three cyclists riding in the foreground, colorful houses of Nyhavn harbor in the background

Copenhagen increased its bicycle budget - I was asked to comment: https://www.verdensmaal.org/nyheder/k%C3%B8benhavns-kommune-vil-g%C3%B8re-det-lettere-at-suse-ge

Photograph of a smiling female researcher

Ditte Bjerregaard has joined NERDS
nerds.itu.dk/2025/10/13/d...

Seems bad:

"Urban highways are associated with decreased social connectivity."

"This barrier effect [is] consistent with historical cases of highways that were built to purposefully disrupt or isolate Black neighborhoods."

doi.org/10.1073/pnas...

Uncovering large inconsistencies between machine learning derived gridded settlement datasets - EPJ Data Science High-resolution human settlement maps provide detailed delineations of where people live and are vital for scientific and practical purposes, such as rapid disaster response, allocation of humanitarian resources, and international development. The increased availability of high-resolution satellite imagery, combined with powerful techniques from machine learning and artificial intelligence (AI), has spurred the creation of a wealth of settlement datasets. The agreement and alignment between these datasets has not been studied in detail. We compare three settlement maps developed by Google (Open Buildings), Meta (High Resolution Population Density Maps) and Microsoft (Global Building Footprints), and uncover which factors drive mismatch. Our study focuses on 44 African countries. We build a global machine learning model to predict where datasets agree, and find that geographic and socio-economic factors considerably impact overlap. However, we also find there is great variability across countries, suggesting complex interactions between country morphology and dataset overlap. It is vital to understand the shortcomings of AI-derived settlement layers as international organizations, governments, and NGOs are already experimenting with incorporating these into programmatic work. We anticipate our work to be a starting point for more critical and detailed analyses of AI derived datasets for humanitarian, policy, and scientific purposes.

New paper out! (well relatively new, I forgot to post about it 🙃)

Have you ever worked with settlement dataset and wondered which one to use, and which are best?

We have!

Joint work w. Andrea Martini, Manuel Garcia-Herranz, and Do-Hyung Kim

epjdatascience.springeropen.com/articles/10....

1/8

How does online communication adapt to organizational context?

In a new pre-print with @cerenbudak.bsky.social and @lajello.bsky.social, we study US labor unions’ Facebook use of discourse frames around wins and losses in representation elections. 🧵

📖 arxiv.org/abs/2510.01757

Gallery of various network visualisations.

🤩 Fantastic new network plotting package available in Python by Fabio Zanini. The package supports both networkx and igraph networks, and has a wide variety of styling options. iplotx.readthedocs.io/en/latest/

Autonorama

Peter Norton - Autonorama islandpress.org/books/autono...

A memes present in a dataset of a paper. This type of image shows a high level of cultural complexity that cannot be fully interpreted by simply looking at the objects represented in the picture.

Three new NERDS publications: Polarization, image-to-text-mapping, and candidate recommendation
nerds.itu.dk/2025/09/30/t...

Estimating affective polarization on a social network Concerns about polarization and hate speech on social media are widespread. Affective polarization, i.e., hostility among partisans, is crucial in this regard as it links political disagreements to hostile language online. However, only a few methods are available to measure how affectively polarized an online debate is, and the existing approaches do not investigate jointly two defining features of affective polarization: hostility and social distance. To address this methodological gap, we propose a network-based measure of affective polarization that combines both aspects – which allows them to be studied independently. We show that our measure accurately captures the relation between the level of disagreement and the hostility expressed towards others (affective component) and whom individuals choose to interact with or avoid (social distance component). Applying our measure to a large-scale Twitter data set on COVID-19, we find that affective polarization was low in February 2020 and increased to high levels as more users joined the Twitter discussion in the following months.

I'm glad to be a coauthor of the excellent Marilena Hohmann for this paper proposing a way to measure affective polarization on social media: https://doi.org/10.1371/journal.pone.0328210

Speaker in front of screen showing colorful images of horse riders.

Seminar room with about 20 people listening to speaker being introduced

Speaker in front of title slide, audience in foreground

It was great to have Paul Smaldino @psmaldino.bsky.social with us today, discussing evolutionary approaches to human culture and society. As a bonus, we learned why we are wearing pants.

The calm before the storm 🤘

Professor portrait: Luca Aiello wants to make the internet a better place On 24 September 2025 at 14:30, Professor Luca Aiello will give his inaugural lecture in Auditorium 0 at the IT University in Copenhagen.

This Wednesday, 24 September 2025 at 14:30, our Professor Luca Aiello @lajello.bsky.social will give his inaugural lecture in Auditorium 0 at the IT University in Copenhagen!

en.itu.dk/About-ITU/Pr...

Portrait photo of a researcher

Prof. Paul Smaldino @psmaldino.bsky.social will join us in Copenhagen on Wed, Sep 24, 12:30-13:15 to present at an exciting data science seminar!

Topic: Why Use An Evolutionary Perspective To Study Culture?

To attend, see: nerds.itu.dk/event/data-s...