---
title: "Segmentation and Processing of German Court Decisions from"
description: "The authors build a cleaned corpus of 251,038 decisions from the 2022-10-18 dump. 14 references."
authors: ["Harshil Darji", "Martin Heckelmann", "Christina Kratsch", "Gerard de Melo"]
year: 2025
venue: "arXiv:2601.01449 · Legal Knowledge and Information Systems, Frontiers in Artificial Intelligence and Applications, Vol. 416, IOS Press, 2025, pp. 276--281"
source: "https://arxiv.org/abs/2601.01449"
doi: "10.3233/faia251597"
arxiv: "2601.01449"
licence: "CC BY-NC-SA 4.0 (arXiv)"
licence_url: "http://creativecommons.org/licenses/by-nc-sa/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/darji-2026-segmentation-processing-german-court-decisions/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/darji-2026-segmentation-processing-german-court-decisions/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2025](https://md.research.elamur.ai/free/en/papers/2025/page/15/) / Segmentation and Processing of German Court Decisions from Open Legal Data

# Segmentation and Processing of German Court Decisions from Open Legal Data

**Authors:** [Harshil Darji](https://md.research.elamur.ai/free/en/authors/harshil-darji/), [Martin Heckelmann](https://md.research.elamur.ai/free/en/authors/martin-heckelmann/), [Christina Kratsch](https://md.research.elamur.ai/free/en/authors/christina-kratsch/) and [Gerard de Melo](https://md.research.elamur.ai/free/en/authors/gerard-de-melo/) · **Year:** 2025 · **Venue:** [arXiv:2601.01449 · Legal Knowledge and Information Systems, Frontiers in Artificial Intelligence and Applications, Vol. 416, IOS Press, 2025, pp. 276--281](https://md.research.elamur.ai/free/en/venues/frontiers-in-artificial-intelligence-and-applications/) ([Frontiers in artificial intelligence and applications 2025](https://md.research.elamur.ai/free/en/venues/frontiers-in-artificial-intelligence-and-applications/2025/)) · **Licence:** [CC BY-NC-SA 4.0 (arXiv)](http://creativecommons.org/licenses/by-nc-sa/4.0/)

**Links:** [Publisher's page](https://doi.org/10.3233/faia251597) · [PDF](https://arxiv.org/pdf/2601.01449) · [DOI 10.3233/faia251597](https://doi.org/10.3233/faia251597) · [arXiv 2601.01449](https://arxiv.org/abs/2601.01449) · [OpenAlex](https://openalex.org/W4416988513)

**Citations:** 1 citations (OpenAlex, 2026-10-14)

## Summary
The Open Legal Data dump of German court decisions has tidy metadata, but the decision bodies are stored as irregular HTML in which the standard parts of a ruling are often not marked. The authors build a cleaned corpus of 251,038 decisions from the 2022-10-18 dump. Each text is divided into Tenor (operative ruling), Tatbestand (facts) and Entscheidungsgründe (reasoning). The Rechtsmittelbelehrung (appeal instructions) goes into its own field.

The pipeline is entirely rule-based:
- visible HTML elements are flattened and whitespace is normalized;
- court location metadata is resolved through the Open Legal Data API;
- section starts are found by exact full-line header matches, including letter-spaced spellings;
- for Beschlüsse, a numbered Gründe section is mapped onto facts and reasoning;
- statute and case citations are attached with an existing extraction tool.

The authors sized a manual check with Cochran's formula and reviewed 384 randomly drawn decisions; 97.40% were segmented correctly. The JSONL release on Hugging Face can support retrieval, citation analysis, rhetorical role classification and section-aware RAG.

## Key points
- The corpus has 251,038 decisions, published as `harshildarji/openlegaldata` (JSONL). Each record holds normalized metadata, sectioned text and extracted law/case references.
- Header detection works as follows:
  - It uses a fixed vocabulary (tenor, tatbestand, entscheidungsgründe, gründe).
  - Each term is matched by two case-insensitive full-line regex forms, one compact and one spaced-letter.
  - Text before the first header defaults to Tenor.
  - In Beschlüsse, Gründe I maps to Tatbestand and Gründe II to Entscheidungsgründe; an undivided Gründe counts entirely as reasoning.
- The verification sample size comes from Cochran's formula with p=0.5, e=0.05 and Z=1.96 (n₀≈384.16), with finite population correction (≈383.58), rounded up to 384. A decision counted as correct only if its sections did not overlap.
- 97.40% of the sample was correct, with a 95% CI of 95.8%–98.9% for the full corpus. The errors came mostly from rare HTML formatting quirks.
- Section coverage:
  - Tenor appears in 87.7% of decisions, Tatbestand in 65.4% and Entscheidungsgründe in 95.1%.
  - All three sections appear in 57.5%, Tenor plus reasoning only in 25.4%, and Tenor alone in 4.5%.
  - The appeal notice appears in 8,335 decisions (3.32%), and 176 decisions have empty content.

## Interesting ideas
- Instead of eyeballing a few outputs, the authors check a rule-based pipeline by drawing a statistically sized random sample, judging each item pass/fail by a strict criterion, and reporting a confidence interval.
- Variant document layouts (Urteil vs. Beschluss, Gründe with or without I/II) are mapped onto one canonical schema, with an explicit fallback rule when the subdivision is missing.
- Procedural boilerplate (the appeal notice) is stored apart from the substantive content, so it does not pollute downstream analysis of the decision itself.
- The section a statement sits in changes what it means: a statute cited in the reasoning carries different weight from one in the operative part, and mixing facts with arguments causes false retrieval matches.

## For Mova

How these ideas could serve Mova, our small open language model — its files, stages and tests:

- **I1** (learn): Silver trees or self-induced repair rules could be audited on ~384 random items with a binary criterion, so each reported accuracy carries a real error bar.
- **I2** (tree): The same pattern suits UD dialect conversion, where each source convention gets a documented mapping plus a stated default for ambiguous cases.
- **I3** (frames): Meta-text such as instructions, morals or editorial notes could be tagged before frame extraction, so no who-did-what frames are built from it.
- **I4** (world): Each world entry could record which part of the text it came from, letting the world model separate narrated facts from claims or arguments when answering.

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

## Institutions

- [University of Potsdam](https://md.research.elamur.ai/free/en/universities/university-of-potsdam/) — University, Potsdam, Germany · 44 works · 5,305 citations
- [Hasso Plattner Institute](https://md.research.elamur.ai/free/en/labs/hasso-plattner-institute/) — Research institute, Potsdam, Germany · 12 works · 4,998 citations
- [HTW Berlin - University of Applied Sciences](https://md.research.elamur.ai/free/en/universities/htw-berlin-university-of-applied-sciences/) — University, Berlin, Germany · 3 works · 1 citation

📚 Shelves (6): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 16,667 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 16,194 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 16,039 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 15,891 works · [🤝 Social Sciences](https://md.research.elamur.ai/free/en/shelves/social-sciences/) · 1,772 works · [🤝 Political Science and International Relations](https://md.research.elamur.ai/free/en/shelves/social-sciences/political-science-and-international-relations/) · 182 works

🧭 Topics (2): [🧮 Reasoning](https://md.research.elamur.ai/free/en/topics/reasoning/) · 1,426 works · [⚖️ Legal AI](https://md.research.elamur.ai/free/en/topics/legal-ai/) · 134 works

## Cite

Cite as: Harshil Darji, Martin Heckelmann, Christina Kratsch and Gerard de Melo. 2025. Segmentation and Processing of German Court Decisions from Open Legal Data. arXiv:2601.01449 · Legal Knowledge and Information Systems, Frontiers in Artificial Intelligence and Applications, Vol. 416, IOS Press, 2025, pp. 276--281.

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

```bibtex
@misc{darji2025segmentation,
  title = {{Segmentation and Processing of German Court Decisions from Open Legal Data}},
  author = {Harshil Darji, Martin Heckelmann, Christina Kratsch and Gerard {de Melo}},
  year = {2025},
  doi = {10.3233/faia251597},
  eprint = {2601.01449},
  archivePrefix = {arXiv},
  url = {https://doi.org/10.3233/faia251597}
}
```

- [← previous SATLab at BAREC Shared Task 2025: Optimizing a Language-Independent System for Fine-Grained Readability Assessment](https://md.research.elamur.ai/free/en/papers/bestgen-2025-satlab-barec-shared-task-2025/)
- [next → SHA at the QIAS Shared Task: LLMs for Arabic Islamic Inheritance Reasoning](https://md.research.elamur.ai/free/en/papers/altammami-2025-sha-qias-shared-task-llms/)
- [↑ back to papers of 2025](https://md.research.elamur.ai/free/en/papers/2025/page/15/)

## References

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

1. Ostendorff M, Blume T, Ostendorff S. Towards an open platform for legal information. In: Proceedings of the ACM/IEEE Joint Conference on Digital Libraries in 2020; 2020. p. 385-8. [find](https://scholar.google.com/scholar?q=Ostendorff+M%2C+Blume+T%2C+Ostendorff+S%2E+Towards+an+open+platform+for+legal+information%2E+In%3A+Proceedings+of+the+ACM%2FIEEE+Joint+Conference+on+Digital+Libraries+in+2020%3B+2020%2E+p%2E+385%2D8%2E)
2. Wrzalik M, Krechel D. GerDaLIR: A German dataset for legal information retrieval. In: Proceedings of the natural legal language processing workshop 2021; 2021. p. 123-8. [find](https://scholar.google.com/scholar?q=Wrzalik+M%2C+Krechel+D%2E+GerDaLIR%3A+A+German+dataset+for+legal+information+retrieval%2E+In%3A+Proceedings+of+the+natural+legal+language+processing+workshop+2021%3B+2021%2E+p%2E+123%2D8%2E)
3. Darji H, Mitrović J, Granitzer M. Exploring Semantic Similarity Between German Legal Texts and Referred Laws. In: International Joint Conference on Knowledge Discovery, Knowledge Engineering, and Knowledge Management. Springer; 2021. p. 37-50. [find](https://scholar.google.com/scholar?q=Darji+H%2C+Mitrovic%CC%81+J%2C+Granitzer+M%2E+Exploring+Semantic+Similarity+Between+German+Legal+Texts+and+Referred+Laws%2E+In%3A+International+Joint+Conference+on+Knowledge+Discovery%2C+Knowledge+Engineering%2C+and+Kno)
4. Darji H, Mitrović J, Granitzer M. Challenges and considerations in annotating legal data: A comprehensive overview. arXiv preprint arXiv:240717503. 2024. [find](https://scholar.google.com/scholar?q=Darji+H%2C+Mitrovic%CC%81+J%2C+Granitzer+M%2E+Challenges+and+considerations+in+annotating+legal+data%3A+A+comprehensive+overview%2E+arXiv+preprint+arXiv%3A240717503%2E+2024%2E)
5. Ma Y, Wu Y, Ai Q, Liu Y, Shao Y, Zhang M, et al. Incorporating structural information into legal case retrieval. ACM Transactions on Information Systems. 2023;42(2):1-28. [find](https://scholar.google.com/scholar?q=Ma+Y%2C+Wu+Y%2C+Ai+Q%2C+Liu+Y%2C+Shao+Y%2C+Zhang+M%2C+et+al%2E+Incorporating+structural+information+into+legal+case+retrieval%2E+ACM+Transactions+on+Information+Systems%2E+2023%3B42%282%29%3A1%2D28%2E)
6. Santosh T, Isaia A, Hong S, Grabmair M. Hiculr: Hierarchical curriculum learning for rhetorical role labeling of legal documents. arXiv preprint arXiv:240918647. 2024. [find](https://scholar.google.com/scholar?q=Santosh+T%2C+Isaia+A%2C+Hong+S%2C+Grabmair+M%2E+Hiculr%3A+Hierarchical+curriculum+learning+for+rhetorical+role+labeling+of+legal+documents%2E+arXiv+preprint+arXiv%3A240918647%2E+2024%2E)
7. Bambroo P, Adhikary S, Bhattacharya P, Chakraborty A, Ghosh S, Ghosh K. MARRO: multi-headed attention for rhetorical role labeling in legal documents. Artificial Intelligence and Law. 2025:1-30. [find](https://scholar.google.com/scholar?q=Bambroo+P%2C+Adhikary+S%2C+Bhattacharya+P%2C+Chakraborty+A%2C+Ghosh+S%2C+Ghosh+K%2E+MARRO%3A+multi%2Dheaded+attention+for+rhetorical+role+labeling+in+legal+documents%2E+Artificial+Intelligence+and+Law%2E+2025%3A1%2D30%2E)
8. Glaser I, Moser S, Matthes F. Improving Legal Information Retrieval: Metadata Extraction and Segmentation of German Court Rulings. In: KDIR; 2021. p. 282-91. [find](https://scholar.google.com/scholar?q=Glaser+I%2C+Moser+S%2C+Matthes+F%2E+Improving+Legal+Information+Retrieval%3A+Metadata+Extraction+and+Segmentation+of+German+Court+Rulings%2E+In%3A+KDIR%3B+2021%2E+p%2E+282%2D91%2E)
9. Urchs S, Mitrović J, Granitzer M. Towards classifying parts of german legal writing styles in german legal judgments. In: 2020 10th International Conference on Advanced Computer Information Technologies (ACIT). IEEE; 2020. p. 451-4. [find](https://scholar.google.com/scholar?q=Urchs+S%2C+Mitrovic%CC%81+J%2C+Granitzer+M%2E+Towards+classifying+parts+of+german+legal+writing+styles+in+german+legal+judgments%2E+In%3A+2020+10th+International+Conference+on+Advanced+Computer+Information+Technolog)
10. Urchs S, Mitrovic J, Granitzer M. Design and Implementation of German Legal Decision Corpora. In: ICAART (2); 2021. p. 515-21. [find](https://scholar.google.com/scholar?q=Urchs+S%2C+Mitrovic+J%2C+Granitzer+M%2E+Design+and+Implementation+of+German+Legal+Decision+Corpora%2E+In%3A+ICAART+%282%29%3B+2021%2E+p%2E+515%2D21%2E)
11. Cochran WG. Sampling techniques. john wiley & sons; 1977. [find](https://scholar.google.com/scholar?q=Cochran+WG%2E+Sampling+techniques%2E+john+wiley+%26+sons%3B+1977%2E)
12. Glaser I, Moser S, Matthes F. Summarization of German court rulings. In: Proceedings of the Natural Legal Language Processing Workshop 2021; 2021. p. 180-9. [find](https://scholar.google.com/scholar?q=Glaser+I%2C+Moser+S%2C+Matthes+F%2E+Summarization+of+German+court+rulings%2E+In%3A+Proceedings+of+the+Natural+Legal+Language+Processing+Workshop+2021%3B+2021%2E+p%2E+180%2D9%2E)
13. Darji H, Mitrović J, Granitzer M. A dataset of German legal reference annotations. In: Proceedings of the Nineteenth International Conference on Artificial Intelligence and Law; 2023. p. 392-6. [find](https://scholar.google.com/scholar?q=Darji+H%2C+Mitrovic%CC%81+J%2C+Granitzer+M%2E+A+dataset+of+German+legal+reference+annotations%2E+In%3A+Proceedings+of+the+Nineteenth+International+Conference+on+Artificial+Intelligence+and+Law%3B+2023%2E+p%2E+392%2D6%2E)
14. Heckelmann M. Smart Contracts. Neue Juristische Wochenschrift (NJW). 2018:504-9. [find](https://scholar.google.com/scholar?q=Heckelmann+M%2E+Smart+Contracts%2E+Neue+Juristische+Wochenschrift+%28NJW%29%2E+2018%3A504%2D9%2E)

*Updated 30 October 2026*