openlegaldata/legal-citation-graph-germany
收藏资源简介：
--- license: odbl dataset_info: - config_name: dump-20260520 features: - name: from_case_court_chamber dtype: string - name: from_case_court_city dtype: string - name: from_case_court_jurisdiction dtype: string - name: from_case_court_level_of_appeal dtype: string - name: from_case_court_name dtype: string - name: from_case_court_state dtype: string - name: from_case_date dtype: date32 - name: from_case_file_number dtype: string - name: from_case_review_status dtype: string - name: from_case_source_name dtype: string - name: from_case_type dtype: string - name: from_law_book_slug dtype: string - name: from_slug dtype: string - name: from_type dtype: string - name: to_case_court_jurisdiction dtype: string - name: to_case_court_level_of_appeal dtype: string - name: to_case_court_name dtype: string - name: to_law_book_code dtype: string - name: to_law_book_slug dtype: string - name: to_law_section dtype: string - name: to_law_title dtype: string - name: to_slug dtype: string - name: to_type dtype: string splits: - name: train num_bytes: 2019159004 num_examples: 7438459 download_size: 72720810 dataset_size: 2019159004 - config_name: dump-20260520-10k features: - name: from_case_court_chamber dtype: string - name: from_case_court_city dtype: string - name: from_case_court_jurisdiction dtype: string - name: from_case_court_level_of_appeal dtype: string - name: from_case_court_name dtype: string - name: from_case_court_state dtype: string - name: from_case_date dtype: date32 - name: from_case_file_number dtype: string - name: from_case_review_status dtype: string - name: from_case_source_name dtype: string - name: from_case_type dtype: string - name: from_law_book_slug dtype: string - name: from_slug dtype: string - name: from_type dtype: string - name: to_case_court_jurisdiction dtype: string - name: to_case_court_level_of_appeal dtype: string - name: to_case_court_name dtype: string - name: to_law_book_code dtype: string - name: to_law_book_slug dtype: string - name: to_law_section dtype: string - name: to_law_title dtype: string - name: to_slug dtype: string - name: to_type dtype: string splits: - name: train num_bytes: 2722440 num_examples: 10000 download_size: 583880 dataset_size: 2722440 - config_name: dump-20260520-1k features: - name: from_case_court_chamber dtype: string - name: from_case_court_city dtype: string - name: from_case_court_jurisdiction dtype: string - name: from_case_court_level_of_appeal dtype: string - name: from_case_court_name dtype: string - name: from_case_court_state dtype: string - name: from_case_date dtype: date32 - name: from_case_file_number dtype: string - name: from_case_review_status dtype: string - name: from_case_source_name dtype: string - name: from_case_type dtype: string - name: from_law_book_slug dtype: string - name: from_slug dtype: string - name: from_type dtype: string - name: to_case_court_jurisdiction dtype: string - name: to_case_court_level_of_appeal dtype: string - name: to_case_court_name dtype: string - name: to_law_book_code dtype: string - name: to_law_book_slug dtype: string - name: to_law_section dtype: string - name: to_law_title dtype: string - name: to_slug dtype: string - name: to_type dtype: string splits: - name: train num_bytes: 271033 num_examples: 1000 download_size: 83828 dataset_size: 271033 configs: - config_name: dump-20260520 data_files: - split: train path: dump-20260520/train-* - config_name: dump-20260520-10k data_files: - split: train path: dump-20260520-10k/train-* - config_name: dump-20260520-1k data_files: - split: train path: dump-20260520-1k/train-* --- # Open Legal Data: Citation Graph Germany This dataset contains the citation graph extracted from German court decisions and laws indexed by the [Open Legal Data Platform](https://de.openlegaldata.io). Each row is a single citation edge from a source (case or law) to a target (case or law), with the slugs needed to join back to the [`court-decisions-germany`](https://huggingface.co/datasets/openlegaldata/court-decisions-germany) dataset. ## Available dumps | Date | Configs | | --- | --- | | 2026-05-20 | `dump-20260520`, `dump-20260520-10k`, `dump-20260520-1k` | ## Data format Each sample has the following columns: - `from_type` — ``Case`` or ``Law`` (source side) - `from_slug` — slug of the citing object - `from_law_book_slug` — book slug when ``from_type == "Law"``; empty otherwise - `from_case_*` — court / file number / date / etc. when ``from_type == "Case"`` - `to_type` — ``Case`` or ``Law`` (target side) - `to_slug` — slug of the cited object - `to_law_book_slug`, `to_law_book_code`, `to_law_section`, `to_law_title` — populated for Law targets - `to_case_court_*` — populated for Case targets Numerical IDs are intentionally omitted — slugs are sufficient and stable across snapshot dates. ## Discord > [!NOTE] > We're back! This project is getting a fresh update - join us on [Discord](https://discord.gg/WCy3aq25ZF) to help revive it. ## Citation If you use this data, please cite our [research paper](https://arxiv.org/abs/2005.13342): ```bibtex @inproceedings{10.1145/3383583.3398616, author = {Ostendorff, Malte and Blume, Till and Ostendorff, Saskia}, title = {Towards an Open Platform for Legal Information}, year = {2020}, isbn = {9781450375856}, publisher = {Association for Computing Machinery}, address = {New York, NY, USA}, url = {https://doi.org/10.1145/3383583.3398616}, doi = {10.1145/3383583.3398616}, booktitle = {Proceedings of the ACM/IEEE Joint Conference on Digital Libraries in 2020}, pages = {385–388}, numpages = {4}, keywords = {open data, open source, legal information system, legal data}, location = {Virtual Event, China}, series = {JCDL '20} } ``` ## License In Germany, acts, statutory instruments, official decrees and official notices, as well as decisions (including court decisions) and official head notes of decisions do not enjoy copyright protection (UrhG § 5). The collection and metadata has the Open Database License (ODbL) v1.0.
This dataset contains the citation graph extracted from German court decisions and laws indexed by the Open Legal Data Platform. Each row is a single citation edge from a source (case or law) to a target (case or law), with the slugs needed to join back to the court-decisions-germany dataset. It includes multiple versions such as dump-20260520 (full), dump-20260520-10k (10k samples), and dump-20260520-1k (1k samples). Columns include from_type (source type, e.g., case or law), from_slug (slug of the citing object), from_case_* columns (court, file number, date, etc. when source is a case), to_type (target type), to_slug (slug of the cited object), to_law_* columns (book information when target is a law), and to_case_court_* columns (court information when target is a case). The dataset is based on German laws that do not enjoy copyright protection and is licensed under the Open Database License (ODbL) v1.0.
