README.md

---
license: cc
language:
- ja
---
# Japanese_OCR

This dataset contains four collections of Japanese government and financial institution PDF documents, intended for OCR research, document layout parsing, and financial text extraction.

## 📂 Contents
| Folder           | Description                                                                 | Source Institution                      | Files | Size  |
|------------------|-----------------------------------------------------------------------------|------------------------------------------|--------|--------|
| `FSA_Whitepaper` | Annual white papers published by the Financial Services Agency (金融庁)       | [FSA](https://www.fsa.go.jp/)            | 331    | 1.5 GB |
| `BOJ_PDFs`       | Research papers and discussion papers from the Bank of Japan (日本銀行)         | [BOJ](https://www.boj.or.jp/)            | 672    | 464 MB |
| `JPX_PDFs`       | Monthly equity market reports from Japan Exchange Group (JPX)               | [JPX](https://www.jpx.co.jp/)            | 2246   | 450 MB |
| `SESC_Report`    | Annual activity reports from the Securities and Exchange Surveillance Commission (SESC) | [FSA - SESC](https://www.fsa.go.jp/sesc/) | 140    | 577 MB |

Total size: ~3 GB

## 🧾 Licensing and Usage

All documents in this dataset are publicly accessible from the official websites of Japanese financial and regulatory authorities. However:

- **JPX PDFs** included in this dataset are publicly available market reports from the official Japan Exchange Group website.  
No content has been modified. This subset is provided **solely for academic, non-commercial OCR benchmarking**.  

- **BOJ PDFs** are discussion and research papers published by the Bank of Japan, typically for academic and policy-oriented dissemination.  
  These are used here under the assumption of fair academic use and should be cited when referenced.

- **FSA Whitepapers** and **SESC Reports** are public information releases from Japan’s Financial Services Agency.  
  These documents are intended for transparency and public policy communication and are considered public-access content under typical non-commercial usage.


## 🔍 Use Cases

This collection can be used for:

- Optical Character Recognition (OCR) benchmarking
- Document layout analysis (tables, headings, multi-column)
- Japanese financial terminology extraction
- Training document parsing models in multilingual or financial settings

## 🏷 Tags

`OCR`, `Japanese`, `PDF`, `financial-documents`, `government-data`, `non-commercial`, `public-sector`