Elamur Research⌕
☰
home / papers / 2022 / Making Italian Parliamentary Records Machine-Actionable: the Construction of the ParlaMint-IT corpus

Making Italian Parliamentary Records Machine-Actionable: the Construction of the ParlaMint-IT corpus

Summary

The authors describe how they built ParlaMint-IT, the Italian component of a multilingual collection of corpora of parliamentary debates within the CLARIN ParlaMint initiative. The corpus covers only the plenary sessions of the Senate of the Italian Republic, from 2013 to November 2020. It is divided into a subcorpus for the COVID-19 pandemic period and a reference part covering the preceding years. The transcripts were taken from the HTML versions, because the Akoma Ntoso format covers only records from 2018 onwards. The embedded markup of speeches and speakers was extracted from the HTML, cleaned, and converted into the ParlaCLARIN TEI XML format. The corpus was enriched with detailed metadata on senators, government members, parties, parliamentary groups and coalitions, taken from the Senate's open data portal. The texts were then automatically annotated, without manual validation, following the UD guidelines: sentences, tokens, POS, lemmas, and dependency syntax. Named entity recognition was added as well. The authors wrote a dedicated converter to transform CoNLL-U into the ParlaMint.ana format. The corpus is comparable with the other national ParlaMint corpora and is therefore suitable for comparative studies of political discourse. It is already being used in research and teaching.

Key points

  • Corpus size: 1199 sessions, 79 373 speeches, approximately 31 million words. Period: from 15.03.2013 to 18.11.2020, i.e. the XVII and XVIII legislatures.
  • The COVID-19 subcorpus begins on 1 November 2019 and contains 115 sessions. The reference part contains 1084 sessions.
  • About 5% of the intermediate XML files (62 of 1199) had to be corrected manually to make them valid XML. Appendices to the transcripts were discarded.
  • Metadata were taken from dati.senato.it (in particular via the SPARQL endpoint). Places of birth were linked to GeoNames. Data on government members who are not senators were entered manually.
  • The corpus is part of ParlaMint 2.1 and ParlaMint.ana 2.1 in the CLARIN repository. NER annotation in IOB format is transferred into TEI <name> elements by means of a two-state automaton.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

Fields: Linguistics & languages

Topics: Named entity recognition

Cited by

6 papers of our library cite it (of the 140673 whose bibliographies we read); 1 of them have a page here, 5 not yet.

Cite

The paper itself, not this page. BibTeX from the library's record of the work.

@inproceedings{agnoloni2022making,
  title = {{Making Italian Parliamentary Records Machine-Actionable: the Construction of the ParlaMint-IT corpus}},
  author = {Agnoloni, Tommaso and Bartolini, Roberto and Frontini, Francesca and Montemagni, Simonetta and Marchetti, Carlo and Quochi, Valeria and Ruisi, Manuela and Venturi, Giulia},
  booktitle = {Proceedings of the Workshop ParlaCLARIN III within the 13th Language Resources and Evaluation Conference},
  year = {2022},
  url = {https://aclanthology.org/2022.parlaclarin-1.17/}
}

References

4 references, 0 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Bański, P., Haaf, S., and Mueller, M. (2018). Lightweight grammatical annotation in the TEI: New perspectives. In Proceedings of the Eleventh International Conference on Language Resources and Evaluation (LREC 2018), Miyazaki, Japan, May. European Language Resources Association (ELRA). find
  2. Del Fante, D. and Cavalieri, A. (2021). Politics in between crises. A political and textual comparative analysis of budgetary speeches and expenditure. In Proceedings of the 1st Workshop on Computational Linguistics for Political Text Analysis. find
  3. Del Fante, D. (2022). ParlaMint – IT – Il corpus del Senato Italiano. Una guida pratica per l’interrogazione del corpus ParlaMint-IT con NoSketch Engine, a supporto dell’analisi del dis-corso politico. https://doi.org/10.5281/ zenodo.6526914. 23 Available from noske/index.HTML 10. find
  4. Language Resource References Bosco, Cristina and Montemagni, Simonetta and Simi, Maria. (2013). Converting Italian Treebanks: Towards an Italian Stanford Dependency Treebank. Association for Computational Linguistics, PID https://aclanthology.org/W13-2308. Erjavec, Tomaž and Ogrodniczuk, Maciej and… ↗

This page in Markdown → https://md.research.elamur.ai/free/en/papers/agnoloni-2022-making-italian-parliamentary-records-machine/