| --- |
| license: gpl-3.0 |
| language: |
| - cs |
| tags: |
| - legal |
| --- |
| ## CzechVOC Dataset |
| The CzechVOC dataset is a structured collection of legal terms and their definitions, curated to support the Czech e-Sbírka system—a digital repository for Czech laws and international treaties. This dataset serves as a comprehensive vocabulary, facilitating consistent referencing and understanding of legal concepts within the Czech legal framework. |
|
|
| ### Dataset Structure |
| The dataset comprises entries with the following columns: |
|
|
| 1. **Index**: A unique identifier for each term. (just counter) |
| 2. **Legal Term**: The specific legal term or concept. |
| 3. **Definition**: A detailed explanation or description of the legal term. |
|
|
| ### Data Collection Methodology |
| The dataset was compiled using a web scraping approach, targeting the e-Sbírka website. The process involved: |
|
|
| Navigating Legal Terms: Accessing pages listing valid legal terms. |
| Extracting Term Details: For each term, retrieving its name and corresponding definition. |
| Data Storage: Organizing the collected data into a structured format with the specified columns. |
|
|
| ### Notes |
| - Please be advised that some definitions are not currently valid, they are only annotated in text with a time span. Only definitions containing starting date are currently valid. |
| - This dataset might be used as precursor for synthetic dataset (but the invalid terms must be filtered) |
|
|
| --- |
| license: gpl-3.0 |
| --- |
