README.md

---
license: cc0-1.0
task_categories:
- token-classification
language:
- en
pretty_name: ArXiv PII
size_categories:
- 1K<n<10K
configs:
- config_name: pages
  data_files:
  - split: "2312"
    path: "data/2312/pages.csv"
  - split: "2401"
    path: "data/2401/pages.csv"
- config_name: labels
  data_files:
  - split: "2312"
    path: "data/2312/labels.csv"
  - split: "2401"
    path: "data/2401/labels.csv"
  default: true
---
# ArXiv PII

Personally Identifiable Information (PII) from ArXiv front pages.

The dataset contains:

- first pages of ArXiv PDFs converted to text
- author related labels: NAME, EMAIL, PHONE

Available splits and data statistics:

|split|pages|NAME|EMAIL|PHONE|
|-----|-----|----|-----|-----|
|**2312**|1749|9871|3993|39|
|**2401**|1312|7477|3039|36|

#### Usage

```python
ds_pages = load_dataset("sorokin/arxiv-pii", "pages", split="2312")
ds_labels = load_dataset("sorokin/arxiv-pii", "labels", split="2312")
```

### Source Data

<!-- This section describes the source data (e.g. news text and headlines, social media posts, translated sentences, ...). -->

- Metadata from Kaggle dataset https://www.kaggle.com/datasets/Cornell-University/arxiv
- PDF files from Google Cloud Storage bucket gs://arxiv-dataset/arxiv/arxiv/pdf/

#### Data Collection and Processing

<!-- This section describes the data collection and processing process such as data selection criteria, filtering and normalization methods, tools and libraries used, etc. -->

1. List of ArXiv papers downloaded from Kaggle dataset `arxiv-metadata-oai-snapshot.json`
2. All papers filtered with **license** equel to `http://creativecommons.org/licenses/by/4.0/`
1. Related PDF files download from GCS
2. First page converted to text with `poppler` utils `pdftotext -l 1 -bbox-layout -nopgbrk`
3. Authors related entities manually tagged in Label Studio