Mapping (A)Ideology : a taxonomy of European parties using generative LLMs as zero-shot learners

Simple item page
dc.contributor.authorDI LEO, Riccardo
dc.contributor.authorZENG, Chen
dc.contributor.authorDINAS, Elias
dc.contributor.authorTAMTAM, Reda
dc.date.accessioned2025-05-05T07:17:17Z
dc.date.available2025-05-05T07:17:17Z
dc.date.issued2025
dc.descriptionPublished online: 14 April 2025
dc.description.abstractWe perform the first mapping of the ideological positions of European parties using generative Artificial Intelligence (AI) as a “zero-shot” learner. We ask OpenAI’s Generative Pre-trained Transformer 3.5 (GPT-3.5) to identify the more “right-wing” option across all possible duplets of European parties at a given point in time, solely based on their names and country of origin, and combine this information via a Bradley–Terry model to create an ideological ranking. A cross-validation employing widely-used expert-, manifesto- and poll-based estimates reveals that the ideological scores produced by Large Language Models (LLMs) closely map those obtained through the expert-based evaluation, i.e., CHES. Given the high cost of scaling parties via trained coders, and the scarcity of expert data before the 1990s, our finding that generative AI produces estimates of comparable quality to CHES supports its usage in political science on the grounds of replicability, agility, and affordability.
dc.description.sponsorshipThis project has received funding from the European Research Council (ERC, POSTNORM, grant agreement No 101088868).
dc.description.sponsorshipThis article was published Open Access with the support from the EUI Library through the CRUI - CUP Transformative Agreement (2023-2025)
dc.format.mimetypeapplication/pdf
dc.identifier.citationPolitical analysis, 2025, OnlineFirst
dc.identifier.doi10.1017/pan.2025.7
dc.identifier.issn1047-1987
dc.identifier.issn1476-4989
dc.identifier.urihttps://hdl.handle.net/1814/92572
dc.language.isoen
dc.orcid.putcode1814/80302:183528837
dc.orcid.putcode1814/81011:183528840
dc.orcid.putcode1814/82762:184401308
dc.publisherCambridge University Press
dc.relationPost-Authoritarian Norms and the Ideological Legacy of Dictatorships
dc.relation.ispartofPolitical analysis
dc.relation.ispartofseries[POSTNORM]
dc.rightsinfo:eu-repo/semantics/openAccess
dc.rights.licenseAttribution 4.0 Internationalen
dc.rights.urihttp://creativecommons.org/licenses/by/4.0/
dc.subjectIdeology scores
dc.subjectComputational methods
dc.subjectExpert opinion
dc.subjectMatched pairs
dc.subjectText-as-data
dc.titleMapping (A)Ideology : a taxonomy of European parties using generative LLMs as zero-shot learners
dc.typeArticleen
dspace.entity.typePublication
person.identifier.orcid0000-0003-4113-7954
person.identifier.orcid0000-0002-2937-3367
person.identifier.orcid0000-0003-2153-6077
person.identifier.other46368
person.identifier.other54156
person.identifier.other31601
person.identifier.other54360
relation.isAuthorOfPublication40a67278-2dc9-4a6e-b8a0-6c35cfe263f1
relation.isAuthorOfPublication836f8a74-4da1-4c49-9129-bba6abb7419e
relation.isAuthorOfPublication86d19cef-2608-401c-9c96-f14d07e8a60d
relation.isAuthorOfPublicationc90981c4-6241-4113-820f-d7c1d287b355
relation.isAuthorOfPublication.latestForDiscovery40a67278-2dc9-4a6e-b8a0-6c35cfe263f1
relation.isProjectOfPublicationf11c68af-90f8-46ee-a3f9-84d597cfee6a
relation.isProjectOfPublication.latestForDiscoveryf11c68af-90f8-46ee-a3f9-84d597cfee6a
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
Mapping_Ideology_2025.pdf
Size:
816.67 KB
Format:
Adobe Portable Document Format
Description:
Full-text in Open Access, Published version
Download
License bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
license.txt
Size:
4.14 KB
Format:
Item-specific license agreed upon to submission
Description:
Download
Collections
SPS Articles