README.md

---
dataset_info:
  features:
  - name: tokens
    sequence: string
  - name: tags
    sequence: int64
  splits:
  - name: train
    num_bytes: 118725876
    num_examples: 88619
  - name: test
    num_bytes: 29511302
    num_examples: 22110
  download_size: 34363806
  dataset_size: 148237178
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
task_categories:
- token-classification
---

# Mountain Names NER Dataset

## Dataset Description
A Named Entity Recognition dataset focused on identifying mountain names in text. 
The dataset contains tokenized text with corresponding NER tags where:
- Tag 1: Mountain name
- Tag 0: Not a mountain name

## Dataset Structure
The dataset contains two main columns:
- `tokens`: List of tokenized words
- `tags`: Corresponding NER tags (0 or 1)

## Example:
```python
{
    'tokens': ['The', 'Everest', 'is', 'the', 'highest', 'peak'],
    'tags': [0, 1, 0, 0, 0, 0]
}
```

## Usage:
```python
from datasets import load_dataset

dataset = load_dataset("Gepe55o/mountain-ner-dataset")

train_data = dataset["train"]
test_data = dataset["test"]
```

## Dataset creation:
 - Source data collected from [NERetrive](https://arxiv.org/pdf/2310.14282) and [Few-NERD](https://arxiv.org/pdf/2105.07464v6) datasets
 - Filtered for mountain-related entities
 - Converted to binary classification (mountain/non-mountain)