|
--- |
|
task_categories: |
|
- token-classification |
|
language: |
|
- it |
|
size_categories: |
|
- 100K<n<1M |
|
--- |
|
## Dataset Overview |
|
|
|
The dataset provided here is a comprehensive collection supporting the model's training and evaluation. It is particularly suited for those looking to work on or improve Italian NER systems. |
|
|
|
### Key Features |
|
|
|
- **Total Texts:** The dataset contains over 110,000 annotated texts, providing a rich resource for training and testing NER models. |
|
- **Unique Labels:** There are a total of 97,351 different labels within the dataset. These labels cover a broad spectrum of named entities, making the dataset invaluable for developing models capable of identifying a wide variety of entities. |
|
- **Annotation Quality:** Annotations have been meticulously verified to ensure high accuracy and reliability for model training and benchmarking. |
|
|
