TWin of Online Social Networks

Informacje na temat projektu

TWON

Identyfikator umowy o grant: 101095095

DOI

10.3030/101095095

Projekt został zamknięty

Data podpisania przez KE 6 Grudnia 2022

Data rozpoczęcia 1 Kwietnia 2023

Data zakończenia 31 Marca 2026

Finansowanie w ramach

• Culture, creativity and inclusive society

Koszt całkowity

€ 2 989 758,13

Wkład UE

€ 2 989 758,13

Koordynowany przez

UNIVERSITEIT VAN AMSTERDAM
Netherlands

CORDIS oferuje możliwość skorzystania z odnośników do publicznie dostępnych publikacji i rezultatów projektów realizowanych w ramach programów ramowych HORYZONT.

Odnośniki do rezultatów i publikacji związanych z poszczególnymi projektami 7PR, a także odnośniki do niektórych konkretnych kategorii wyników, takich jak zbiory danych i oprogramowanie, są dynamicznie pobierane z systemu OpenAIRE .

Rezultaty

Documents, reports (8)

Deep Learning Report

Report about deep learning topologies and components

Report about strategies to improve platform mechanics
Report on Simulation Model

Report including:1) central model ingredients and their machine-learning ready implementation, including modelling elements, dynamics and network structures;2) simulation results under varying parameters of the calibrated TWON

Impact assessment and ethical/legal guidance handbook

Report about ethical and legal regulatory standards. Including an ethical and legal guidance handbook based on an integrated ethical, legal, and socio-economic impact assessment of TWONs and

Citizen Lab report and materials

Report about Citizen Labs results including communication activities for citizens. Includes, as an appendix, one pagers and presentations used at the Citizen Labs workshops

Field Study Report

Report on outcomes of all field studies. Includes the design and results of both the initial and the final case studies.

Definition of Metrics

Report on the defintion and operationalization of metrics to assess how constructive democratic debates are.

Computational Report

Report of computational aspects of TWON implementation and execution

Other (1)

TWON software platform

TWON software platform: efficient and modular implementaton of models and algorithms developed in the project as well as visualization and analysis

Data sets, microdata, etc (2)

Final data set DS1-DS4

Final curated data set DS1 - DS4.In this release, the data from the initial curated data are prepared for public release. Data sources DS3 (data donation) and DS4 (simulation results) are prepared for public release and added.

ML dataset

Machine learning data set; collected and enriched; prepared for public release

Demonstrators, pilots, prototypes (2)

Prototype of Calibrated TWON

Prototype of calibrated TWON with machine learned parametrization

General TWON prototype

General TWON prototype after the second iteration of development

Websites, patent fillings, videos etc. (2)

Policy handouts

Policy handouts and presentations with key findings are prepared. This feedback to policy makers can also entail engagement with the other projects from our cluster.

Website

Public Project Website set up and released, using a .eu domain and the project acronym as part of the domain name, including a finalized version of the project logo.

Publikacje

Conference proceedings (15)

Zero-shot prompt-based classification: topic labeling in times of foundation models in German Tweets

Autorzy: Simon Münker, Kai Kugler, Achim Rettinger
Opublikowane w: Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 4: Student Research Workshop), 2025
Wydawca: Association for Computational Linguistics
DOI: 10.18653/V1/2025.ACL-SRW.4

Towards Simulating Social Media Users with LLMs: Evaluating the Operational Validity of Conditioned Comment Prediction

Autorzy: Schwager N.; Münker, S.; Plum A; Rettinger A
Opublikowane w: The Proceedings for the 15th Workshop on Computational Approaches to Subjectivity, Sentiment & Social Media Analysis (WASSA 2026), 2026
Wydawca: Association for Computational Linguistics

Don’t Trust Generative Agents to Mimic Communication on Social Networks Unless You Benchmarked their Empirical Realism

Autorzy: Simon Münker, Nils Schwager, Achim Rettinger
Opublikowane w: Proceedings of the 19th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Papers), 2026
Wydawca: Association for Computational Linguistics
DOI: 10.18653/V1/2026.EACL-LONG.51

Emergent Behaviors from LLM-Agent Simulations

Autorzy: Adrian Mladenic Grobelnik,Faizon Zaman, Jofre Espigule-Pons, Marko Grobelnik
Opublikowane w: Data Mining and Data Warehouses - SiKDD : Information Society - IS 2023 : proceedings of the 26th international multiconference, 2023, ISSN 2630-371X
Wydawca: SiKDD : Information Society - IS 2023

Classification of Sparse and Irregularly Sampled Time Series with Convolutional Neural Networks

Autorzy: Buza, Krisztian Antal
Opublikowane w: 2023
DOI: 10.5281/zenodo.8186573

Fact Manipulation in News: LLM-Driven Synthesis and Evaluation of Fake News Annotation

Autorzy: Luka Golob, Abdul Sittar
Opublikowane w: Data Mining and Data Warehouses - SiKDD : proceedings of the 28th International Multiconference, Numer C, 2024, ISSN 2630-371X
Wydawca: Information Society 2024
DOI: 10.70314/IS.2024.SIKDD.13

Refining Deliberative Standards for Online Political Communication: Introducing a Summative Approach to Designing Deliberative Recommender Systems

Autorzy: Sjoerd B. Stolwijk, Corinna Oschatz, Michael Heseltine, Damian Trilling
Opublikowane w: Proceedings of the First Workshop on the Normative Design and Evaluation of Recommender Systems (NORMalize 2023) co-located with the 17th ACM Conference on Recommender Systems (RecSys 2023), 2023, ISSN 1613-0073
Wydawca: CEUR

Towards AI-powered real-time negotiation agent

Autorzy: Martin Žust, Marko Grobelnik, Adrian Mladenić Grobelnik, Abdul Sittar
Opublikowane w: Joint proceedings of the ESWC 2025 Workshops and Tutorials co-located with 22nd Extended Semantic Web Conference (ESWC 2025), Numer 3977, 2025, ISSN 1613-0073
Wydawca: ESWC 2025

An approach to creating a time-series dataset for news propagation: Ukraine-war case study

Autorzy: Abdul Sittar, Dunja Mladenić
Opublikowane w: Data Mining and Data Warehouses - SiKDD : Information Society - IS 2023, Numer 9, 2023, ISSN 2630-371X
Wydawca: SiKDD : Information Society

Time Series Forecasting with Distortion-Aware Convolutional Neural Networks

Autorzy: Buza, Krisztian Antal
Opublikowane w: 2023
DOI: 10.5281/zenodo.8186596

Perception of AI in Slovenia

Autorzy: Abdul SITTAR, Alenka GUČEK, Dunja MLADENIĆ
Opublikowane w: Data Mining and Data Warehouses - SiKDD : Information Society - IS 2023 : proceedings of the 27h international multiconference, Numer C, 2024, ISSN 2630-371X
Wydawca: Information Society - IS - SiKDD 2024
DOI: 10.70314/IS.2024.SIKDD.14

Fingerprinting LLMs through Survey Item Factor Correlation: A Case Study on Humor Style Questionnaire

Autorzy: Simon Münker
Opublikowane w: Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, 2025
Wydawca: Association for Computational Linguistics
DOI: 10.18653/V1/2025.EMNLP-MAIN.13

Guardrail Vulnerabilities in Open-Source Language Models: Implications for Democratic Discourse and Marginalized Communities

Autorzy: Simon Münker, Fabio Sartori
Opublikowane w: Proceedings of the Annual Hawaii International Conference on System Sciences, Proceedings of the 59th Hawaii International Conference on System Sciences, 2026
Wydawca: Hawaii International Conference on System Sciences
DOI: 10.24251/HICSS.2026.804

Demokratiegefährdende Plattform-Mechanismen–Erkennen, Verstehen, Bekämpfen.

Autorzy: Bezzaoui, Isabel; Nevena Nikolajevic, and Jonas Fegert
Opublikowane w: KI| Konflikte| Konventionen–PolKomm2023, 2023
Wydawca: Karlsruher Institut für Technologie (KIT)
DOI: 10.5445/IR/1000161914

Cultural Bias in Large Language Models: Evaluating AI Agents through Moral Questionnaires

Autorzy: Simon Münker
Opublikowane w: Proceedings of 0th Moral and Legal AI Alignment Symposium, 2025
Wydawca: Joint International Association for Computing and Philosophy (IACAP) and the Society for the Study of Artificial Intelligence and Simulation of Behaviour (AISB) Conference on Philosophy of Computing and AI (IACAP/AISB-25)

Peer reviewed articles (10)

Integrating European contexts and needs into WHO guiding principles on online mental health content for young people: five recommendations.

Autorzy: Helena Ludwig-Walz, Martin Bujard, Jonas Fegert & Jörg M. Fegert
Opublikowane w: European Child & Adolescent Psychiatry, 2024, ISSN 1435-165X
Wydawca: Springer
DOI: 10.1007/s00787-024-02566-9

AI alignment: Assessing the global impact of recommender systems.

Autorzy: Bojic, Ljubisa
Opublikowane w: Futures, Numer 160, 2024, ISSN 0016-3287
Wydawca: Elsevier Ltd.
DOI: 10.1016/j.futures.2024.103383

Does GPT-4 surpass human performance in linguistic pragmatics?

Autorzy: Ljubiša Bojić; Predrag Kovačević; Milan Čabarkapa
Opublikowane w: Humanities and Social Sciences Communications, 2025, ISSN 2662-9992
Wydawca: Springer
DOI: 10.48550/ARXIV.2312.09545

Political Bias in LLMs: Unaligned Moral Values in Agent-centric Simulations

Autorzy: Simon Münker
Opublikowane w: Journal for Language Technology and Computational Linguistics, 2025, ISSN 2190-6858
Wydawca: German Society for Computational Linguistics and Language Technology (GSCL)
DOI: 10.48550/ARXIV.2408.11415

CERN for AI: a theoretical framework for autonomous simulation-based artificial intelligence testing and alignment

Autorzy: Ljubiša Bojić, Matteo Cinelli, Dubravko Ćulibrk & Boris Delibašić
Opublikowane w: European Journal of Futures Research, 2024, ISSN 2195-2248
Wydawca: Springer Nature
DOI: 10.1186/s40309-024-00238-0

Profiling the barriers to the spreading of news using news headlines

Autorzy: Abdul Sittar, Dunja Mladenić, Marko Grobelnik
Opublikowane w: Frontiers in artificial intelligence, Numer 6, 2023, ISSN 2624-8212
Wydawca: Frontiers
DOI: 10.3389/frai.2023.1225213

Can we use automated approaches to measure the quality of online political discussion? How to (not) measure interactivity, diversity, rationality, and incivility in online comments to the news

Autorzy: Sjoerd B. Stolwijk, Mark Boukes, Wang Ngai Yeung, Yufang Liao, Simon Münker, Anne C. Kroon, Damian Trilling
Opublikowane w: Communication Methods and Measures, Numer 20, 2025, ISSN 1931-2458
Wydawca: Informa UK Limited
DOI: 10.1080/19312458.2025.2553300

The dual impact of virtual reality: examining the addictive potential and therapeutic applications of immersive media in the metaverse

Autorzy: Ljubiša Bojić, Jörg Matthes, Agariadne Dwinggo Samala, Milan Čabarkapa
Opublikowane w: Information, Communication & Society, Numer 29, 2025, ISSN 1369-118X
Wydawca: Informa UK Limited
DOI: 10.1080/1369118X.2025.2520005

An Agent‑Based Simulation of Politicized Topics Using Large Language Models: Algorithmic Personalization and Polarization on Social Media

Autorzy: Ljubiša Bojić; Velibor Ilić; Veljko Prodanović; Vuk Vuković
Opublikowane w: Chinese Political Science Review, 2025, ISSN 2365-4252
Wydawca: Springer
DOI: 10.1007/S41111-025-00326-X

Digital Democracy: A Wake-Up Call – How IS Research Can Contribute to Strengthening the Resilience of Modern Democracies

Autorzy: Weinhardt, C.; Fegert, J.; Hinz, O.; Aalst, W. M. P. van der
Opublikowane w: Business & Information Systems Engineering, 2024, ISSN 2363-7005
Wydawca: Springer
DOI: 10.1007/S12599-024-00862-X

Monographic books (1)

Navigating the Digital Age. An In-Depth Exploration into the Intersection of Modern Technologies and Societal Transformation

Autorzy: Bojić, Ljubiša; Žikić, Simona; Matthes, Jörg; Trilling, Damian
Opublikowane w: 2024
Wydawca: Belgrade: Institute for Philosophy and Social Theory

Other (1)

Website

Wydawca: FZI