Domain-Specific Automated Style Checking for Legislative Texts

Drafts of legislative texts have to go through several editorial cycles before they are published. The texts not only have to comply with formal and legal aspects but also with domain-specific linguistic style guidelines. Manual assessment of the compliance of legislative texts with these guidelines is extremely time-consuming and error-prone. At the Department of Computational Linguistics, we develop methods to automate this process: we investigate how violations of style guidelines can be detected automatically and corresponding warnings be generated for the editor.

The project focuses on German-language legislative editing in Switzerland.

Zoom (JPG, 393 KB)

Example output of the style checker.

Researchers:

• Stefan Höfler
• Kyoko Sugisaki

Scientific Mentoring:

• Michael Hess (Department of Computational Linguistics)
• Felix Uhlmann (Institute of Law)
• Rebekka Bratschi (Swiss Federal Chancellery)

The project was funded by the Swiss National Science Foundation (05/2011 – 02/2015) and the Forschungskredit of the University of Zurich (08.2014 – 07.2015).

Entry in the SNF research database: Link.

Höfler, Stefan; Sugisaki, Kyoko (2014). Constructing and exploiting an automatically annotated resource of legislative texts Ninth International Conference on Language Resources and Evaluation (LREC'14), Reykjavik.

Sugisaki, Kyoko (2013). Hybrid Morphosyntactic Disambiguation for Parsing Swiss Law Texts in German 35. Jahrestagung der Deutschen Gesellschaft für Sprachwissenschaft (DGfS), Potsdam.

Höfer, Stefan (2012). Automated Style Checking for Swiss Laws Third International Workshop on Controlled Natural Language (CNL 2012), Zürich.

Sugisaki, Kyoko (2014). Automatic detection of syntax-related style violations in law texts, Universität Zürich, Zürich. [Slides]

Höfler, Stefan (2014). «Ein Satz – eine Aussage»: Umsetzung in der Gesetzesredaktion. Jahrestagung 2014 der Vereinigung für Angewandte Linguistik in der Schweiz (VALS), Lugano. [Slides]

Sugisaki, Kyoko (2013). Verbal morphosyntactic disambiguation through topological field recognition in German-language law texts. Third International Workshop on Systems and Frameworks for Computational Morphology (SFCM 2013), Berlin. [Slides]

Höfler, Stefan (2013). Ein Satz – eine Aussage / ein Artikel – eine Norm. Invited talk, Redaktionstag des deutschen Sprachdienstes der Schweizerischen Bundeskanzlei, Guggisberg.

Sugisaki, Kyoko (2013). Incremental Morphosyntactic Disambiguation of Nouns in German-Language Law Texts. ESSLLI-13 Workshop on Extrinsic Parse Improvement (EPI) , Düsseldorf. [Slides]

Höfler, Stefan (2013). Einführung in die maschinelle Textstrukturanalyse. Invited talk, Seminar "Mensch-Computer-Kommunikation: Einführung in die Computerlinguistik", Universität Basel, Basel.

Sugisaki, Kyoko (2013). Morphosyntaktische Desambiguierung von Nomen in Schweizer Gesetzestexten. Kolloquium in Computerlinguistik, Universität Zürich, Zürich. [Slides]

Höfer, Stefan (2012). Legislative Drafting Guidelines: How Different Are They From Controlled Language Rules for Technical Writing? Third International Workshop on Controlled Natural Language (CNL 2012), Zürich. [Slides]

Sugisaki, Kyoko (2012). Challenges in Automating Style Checking for Legislative Texts. EACL 2012 Workshop on Computational Linguistics and Writing, Avignon. [Slides]

Sugisaki, Kyoko (2012). Modellierung von Verletzungen syntaktischer Stilregeln in Gesetzestexten. Kolloquium in Computerlinguistik, Universität Zürich, Zürich. [Slides]

Höfler, Stefan (2012). Stilprüfung und Rechtstextanalyse. Kolloquium in Computerlinguistik, Universität Zürich, Zürich. [Slides]

Höfler, Stefan; Piotrowski, Michael (2011). Building Corpora for the Philological Study of Swiss Legal Texts. Workshop on Annotation of Corpora for Research in the Humanities (ACRH 2012), Heidelberg, Germany.

Höfler, Stefan (2011). Erkennung von Stilrichtlinienverletzungen in Gesetzestexten. Linguistisches Forschungskolloquium, Universität Zürich, Zürich. [Slides]

Höfler, Stefan (2011). Erkennung von Richtlinienverletzungen in Gesetzesentwürfen. Kolloquium in Computerlinguistik, Universität Zürich, Zürich. [Slides]

Höfler, Stefan; Bünzli, Alexandra (2010). Computerlinguistische Methoden für die Gesetzesredaktion. invited talk, Sprachentag 2010, Schweizerische Bundeskanzlei, Bern. [Slides]