HERITAGE: An End-to-End Web Platform for Processing Korean Historical Documents in Hanja

TL;DR

HERITage addresses the bottleneck in translating Korean historical documents written in Hanja by providing an end-to-end web platform with PR, NER, MT, and an interactive glossary. It combines a containerized microservice architecture, a Remix frontend, BERT-based sequence labeling, and vLLM-based translation to deliver rapid predictions and editable outputs that mirror expert workflows. The contributions include a fully open-source toolkit, an integrated three-task pipeline, and tri-lingual glossary support, with substantial performance metrics on standard historical corpora (AJD, KLC). This platform democratizes access to decades of cultural heritage by enabling researchers and the public to read, annotate, and translate Hanja documents, while inviting community feedback and revision of outputs.

Abstract

While Korean historical documents are invaluable cultural heritage, understanding those documents requires in-depth Hanja expertise. Hanja is an ancient language used in Korea before the 20th century, whose characters were borrowed from old Chinese but had evolved in Korea for centuries. Modern Koreans and Chinese cannot understand Korean historical documents without substantial additional help, and while previous efforts have produced some Korean and English translations, this requires in-depth expertise, and so most of the documents are not translated into any modern language. To address this gap, we present HERITAGE, the first open-source Hanja NLP toolkit to assist in understanding and translating the unexplored Korean historical documents written in Hanja. HERITAGE is a web-based platform providing model predictions of three critical tasks in historical document understanding via Hanja language models: punctuation restoration, named entity recognition, and machine translation (MT). HERITAGE also provides an interactive glossary, which provides the character-level reading of the Hanja characters in modern Korean, as well as character-level English definition. HERITAGE serves two purposes. First, anyone interested in these documents can get a general understanding from the model predictions and the interactive glossary, especially MT outputs in Korean and English. Second, since the model outputs are not perfect, Hanja experts can revise them to produce better annotations and translations. This would boost the translation efficiency and potentially lead to most of the historical documents being translated into modern languages, lowering the barrier on unexplored Korean historical documents.

Figures (5)

• Figure 1: Overview of HERITage.
• Figure 2: System architecture diagram.
• Figure 3: Web interface for Punctuation Restoration.
• Figure 4: Web interface for Named Entity Recognition.
• Figure 5: Web interface for Machine Translation.