task_categories:
- token-classification
task_ids:
- named-entity-recognition
language:
- uk
pretty_name: UK-NER-contracts
Dataset Description
Legal Contracts Dataset for Training NER Model This repository contains a specially curated dataset consisting of legal contracts. It is designed for the purpose of training a Named Entity Recognition (NER) model, with the aim to recognize and classify four types of entities in the text:
Contract Type, Clause Title, Clause Number, Definition Title
The dataset includes a broad variety of legal contracts, covering diverse domains such as employment, real estate, services, sale, lease, etc.
Entities in the text have been manually labeled by experts in the field, ensuring high-quality training data for the model.
Each document in the dataset has been annotated in the following format:
(Start_Position, End_Position, Entity_Label)
For example, a clause title may be annotated as follows: (102, 115, 'clause title')
This will assist the NER model in identifying not only the text of the entity, but also its position within the document.
Usage Guidelines