VietAlphaLabs/hoang-xuan-han-dictionary

VietAlpha Hoàng Xuân Hãn Scientific Dictionary

Danh-từ khoa-học (Vocabulaire Scientifique), 1942

VietAlpha Lab · Website

The VietAlpha Hoàng Xuân Hãn Scientific Dictionary is a machine-readable, verified digital edition of Danh-từ khoa-học (Vocabulaire Scientifique: Toán, Lý, Hóa, Cơ, Thiên-văn), compiled by mathematician, historian, and linguist Hoàng Xuân Hãn (黃春瀚, 1908–1996).

First published in Hanoi in 1942 and re-issued in Paris by Nhà xuất-bản Minh-Tân in 1951, the work laid the foundation for modern Vietnamese scientific terminology across Mathematics, Physics, Chemistry, Mechanics, and Astronomy. This release contains 5,783 clean, cross-verified French–Vietnamese entries digitized from the Bibliothèque nationale de France (BnF / Gallica) preservation scans of the 1951 edition.

---

Scientific Domains

Each term may carry one or more domain codes, printed in parentheses after the grammatical label:

Code	Vietnamese Domain	French Equivalent	Field
t	Toán-học	Mathématiques	Mathematics
l	Lý-học	Physique	Physics
h	Hóa-học	Chimie	Chemistry
c	Cơ-học	Mécanique	Mechanics
th	Thiên-văn	Astronomie	Astronomy
k	n/a	n/a	Minerals and rocks (e.g. AGATE, ALBÂTRE, AMÉTHYSTE)

Multi-domain terms use comma-separated codes ("t, l, h"). General vocabulary with no domain marking has domain: null.

---

Record Schema

{
  "entry_id": "e00001",
  "headword": "ABAISSEMENT",
  "grammatical_label": "nm",
  "domain": "t, l, h",
  "content": "1—Sự, cách hạ. Abaissement d’une perpendiculaire: sự, cách hạ một đường thẳng góc. Abaissement du point de fusion: sự hạ dung-điểm. 2—Độ hạ. Abaissement moléculaire: độ-hạ phân-tử."
}

Field	Type	Description
entry_id	str	Stable entry identifier
headword	str	French lemma, uppercase as printed
grammatical_label	str | null	Part of speech as printed: nm, nf, adj, v, adv, pr, np
domain	str | null	Domain code(s) as printed (see table above)
content	str	Vietnamese translation, numbered senses, and collocations/examples

---

Files

File	Description
dictionary.parquet	5,783 entries in Apache Parquet; backs the dictionary config and the Hub viewer
dictionary.jsonl	The same 5,783 entries as JSON Lines (UTF-8, one object per line)
stats.json	Release statistics, page ranges, and pipeline counts
NOTICE	Attribution and historical source statement
LICENSE	Apache License 2.0

---

Loading the Data

Hugging Face datasets

from datasets import load_dataset

dataset = load_dataset(
    "VietAlphaLabs/hoang-xuan-han-dictionary",
    "dictionary",
    split="train",
)

print(f"Loaded {len(dataset)} entries")
print(dataset[0])

Pandas

import pandas as pd

df = pd.read_parquet("dictionary.parquet")

# Entries tagged Mathematics (code "t"), including multi-domain entries
codes = df["domain"].fillna("").str.split(",").apply(lambda cs: [c.strip() for c in cs])
math_terms = df[codes.apply(lambda cs: "t" in cs)]
print(f"Found {len(math_terms)} mathematics entries")

---

Licensing & Attribution

• Transcriptions & dataset code: Licensed under the Apache License, Version 2.0 © 2026 VietAlpha Lab.
• Historical work: The underlying lexical work Danh-từ khoa-học (1942) by Hoàng Xuân Hãn is in the public domain.
• When citing or redistributing this dataset, please retain the NOTICE file and cite as follows:

@dataset{vietalpha2026hoangxuanhan,
  author       = {VietAlpha Lab},
  title        = {VietAlpha Hoàng Xuân Hãn Scientific Dictionary (Danh-từ khoa-học 1942)},
  year         = {2026},
  publisher    = {Hugging Face},
  howpublished = {\url{https://huggingface.co/datasets/VietAlphaLabs/hoang-xuan-han-dictionary}},
  note         = {Digitized and reconstructed from Danh-từ khoa-học by Hoàng Xuân Hãn (1942/1951)}
}