Risultati finali
Text storage module based on elastic research with role and project-based access to metadata and raw text, published via github and the project website.
Project websiteBuilding up the project webpage for all stakeholders - for internal and public use - on basis of conceptual project design.
Online learning material & training sessionIn collaboration with WP9, WP3 online learning material for (first-time) users of the cre-ated media data inventories will be developed.The legal issues related to data scraping are discussed with WP10 and reflected in the types of collected tools and provided learning material.
Web-based annotation moduleWeb-based annotation module to allow for coding of validation and training data, published via github and the project website.
Collection of text preprocessing moduleCollection of text preprocessing module based on state of the art off the shelf open source linguistic tools, published via github and the project website.
Public access webpageWebsite that allows basic search functionality and generation of visual, descriptive analyses (term or topic frequencies) across countries, time, and political actors.
Website: texts political organizationsTo foster the usage of the texts as data sources for the analysis of political organizations a website offering information on both the different text types produced by political organizations and where and how they can be accessed will be set up.
Research toolkitDevelopment of a toolkit for non-consumptive research, meaning that third parties have sufficient access to underlying data to check or replicate studies and conduct new research, but without getting direct access to data that cannot be publicly shared due to copyright, privacy, or other concerns.
Tool for data linkageDevelop a tool that facilitates and partly automatizes data linkages. A tool (for example in R, or Py-thon) will be build. Based on substantial choices made by the researchers, the tool facilitates combin-ing datasets.
WP3 coordinates with WP8 that the inventory can be linked to text sources from WP2, WP4, and WP5. Relevant for WP3 are here for example the definition of Europe and the detailed organization of the inventory (e.g., classification of political parties and the categories for political leanings of media sources; labeling and formatting of publication date, country and language codes/abbreviations).
Peer reviewed article IIPeer reviewed article describing the need for non-consumptive research and the functionality of the toolkit.
Evaluation of accessibility & usability of existing text collectionsBased on the inventory an evaluation of the accessibility and usability of existing text collections will be conducted to define best practices for making textual data easily accessible for a variety of users.
Feasibility reportBased on the inventory and the case study the final feasibility report of the infrastructure set up in the context of OPTED will be made.
Review article about media classification typologiesThe results of the conceptual overview of text data are summarized in a review article about journalistic massmediated text types in Europe
Tool collectionProvide a tool collection of a) data scrapers for media websites (and archives, social network APIs, RSS-feeds), of b) data selection methods to select ‘political media texts’ (in accordance with the defi-nition of a political media text), and of c) media text deduplication methods (account for dynamic update of news websites, multiple versions per archive and across archives).
3rd monitoring and observance reportThird monitoring and observance report on the ELSA requirements. A report on the validation of the ELSA requirements after the completion of the final infrastructure design.
Report detailing a proto-type platformA prototype of the platform to house curated inventory. The deliverable will be in the form of a report detailing the platform with an analysis of initial expert and user feedback on the prototype platform.
Systematic literature review of theoretical contextualization of CPPTA description of existing theoretical approaches to CPPT Within a systematic literature review we will build an extensive theoretical contextualization what theoretical approaches have been used to comprehend and explain CPPT over the last 15 years
Workflow linking types of data sourcesDevelop a stepwise workflow for linking different types of data sources, addressing appropriate identifiers and aggregation levels.
Proofs of conceptProofs of concept for selected key validation challenges on the road map, formulating future-oriented best practices, so as to enable a systematic validation of existing approaches, and development of innovative approaches.
Meeting bureaucrats/data scientistsTowards a better infrastructure of parliamentary archives for scientific research Exchange with parliamentary practitioners and data scientists in EU and national parliaments on parliamentary APIs for political science research
Inventory (Database) of media sourcesWithin the decided scope and organisation of the inventory concrete media sources will be listed.
Inventory of data sources & linkageEstablish an inventory of data sources and opportunities for linkage
Training tutorials for ParlLawSpeechDevelop training material for integrated database ParlLawSpeech: easy-to-use sample materials and hands-on instructions for researchers employing our data in a wide variety of research settings and for different research agendas. We provide unhindered data access by contributing training modules, tutorials, and a test framework of a generic API to WP 7 and 9.
Inventory of CPPT data, tools, and methodologyBuilding up an extensive inventory of a types of texts in CPPT and how they are produced b the predominant strategies for collecting CPPT and 3 the varieties of tools and methodological approaches deployed to analyze CPPT
CPPT GuidelinesGuidelines for good practice and proposal for developing a data repository of CPPT in collaboration with WP10 and WP9.
ELSA by design questionnaireQuestionnaire to be shared with partners to gather information about the architecture and functioning of this project
Conceptual frameworkCreation of a framework of common access-related issues and an updated list of specific aspects of CTTP research that might be inhibited by existing legal considerations. The Task will feed into WP10 which will use this map to investigate specific legal frameworks of relevance related both to national frame-works and those associated with the technology platforms such as Facebook, Google and Twitter.
Blueprint on case studyConduct a case study that includes both linkage of data sources as well as concepts. Multiple datasets will be selected and combined to (a) outline the linkage process; (b) demonstrate the usability of the tool and answers substantially relevant questions.
Integrated standard recommendationsFormulate integrated standards and generalized procedures suitable to engage in a systematic comparative validation of multilingual text analysis tools and approaches; collate reference data sets, databases (e.g., word embeddings) and computational resources needed for the implementation of these standards.
Inventory of text collectionsAn overview of available collections of texts for all EU member states, Iceland, Israel, Norway, and Switzerland will be provided.
Review of available parliamentary corporaSystematic overview of available parliamentary speech corpora, online report
Report challenges & solutions for studiesA report will be produced that will identify challenges one encounters when working with such different text types in a comparative analysis and that will develop possible solutions to these challenges.
Peer reviewed article IPeer reviewed article describing the storage system and preprocessing facilities
Update of the CPPT inventoryUpdate of the data and theory inventory review for the duration of the project. Furthermore both repository would be comprehensively described.
Description of text types produced by political organizationsA detailed description of the different types of texts produced by political organizations interesting for political science research will be created.
Creation of ParlLawSpeech parliamentary speech database for national parliaments and the EU. This task builds on and systematically extends the initial experiences and resources developed by the ParlSpeech project (Rauh et al., 2017).
Integrated hub and knowledge base for multilingual computational text analysisBuild a living hub (in the form of a Wiki, whose initial contents are developed by WP6, but with the aim of attracting future contributions from the research community) as knowledge base and central access point for the research community supporting the acquisition, integration and future develop-ment of methodological insights, computational tools and validation resources.
Pubblicazioni
Autori:
Christoph Ivanusch; Lisa Zehnter; Tobias Burst
Pubblicato in:
Electoral Studies, Numero Volume 86, 2023, ISSN 0261-3794
Editore:
Pergamon Press Ltd.
DOI:
10.1016/j.electstud.2023.102703
Autori:
Sebők, M., Proksch, S.-O., Rauh, C., Visnovitz, P., Balázs, G., & Schwalbach, J.
Pubblicato in:
International Political Science Review, 2023, ISSN 1460-373X
Editore:
SAGE Publications
DOI:
10.1177/01925121231199904
Autori:
Rebeka Kiss; Miklós Sebők
Pubblicato in:
International Journal of Parliamentary Studies, Numero 2(2), 2022, Pagina/e 278-284, ISSN 2352-7072
Editore:
Brill
DOI:
10.1163/26668912-bja10053
Autori:
Zachary Greene; Thomas Schober; Thomas Scotto; Graeme Roy
Pubblicato in:
National Institute Economic Review, Numero 1(13), 2023, ISSN 0027-9501
Editore:
SAGE Publications
DOI:
10.1017/nie.2023.23
Autori:
Zachary D Greene; Amanda A Licht
Pubblicato in:
Conflict Management and Peace Science, 2023, ISSN 0738-8942
Editore:
SAGE Publications
DOI:
10.1177/07388942231195300
Autori:
Mariken A.C.G van der Velden; Martijn Schoonvelde2; Christian Baden
Pubblicato in:
Computational Communication Research, Numero Volume 5, Numero 2, 2023, Pagina/e 1, ISSN 2665-9085
Editore:
Amsterdam University Press
DOI:
10.5117/ccr2023.2.1.vand
Autori:
Christian Baden; Christian Pipal; Martijn Schoonvelde; Mariken A. C. G van der Velden
Pubblicato in:
Communication Methods and Measures, Numero 16, 2021, Pagina/e 1-18, ISSN 1931-2458
Editore:
Routledge
DOI:
10.1080/19312458.2021.2015574
Autori:
Bernhard Weßels
Pubblicato in:
The Changing German Voter, Numero 21 Apr. 2022, 2022, Pagina/e 50–77, ISBN 9780191882197
Editore:
online edn, Oxford Academic
DOI:
10.1093/oso/9780198847519.003.0003
È in corso la ricerca di dati su OpenAIRE...
Si è verificato un errore durante la ricerca dei dati su OpenAIRE
Nessun risultato disponibile