This project builds a comprehensive AI research infrastructure for Japanese documents. We will integrate Japanese translations and secondary literature into the machine-translation and search platform Dharmamitra (2026–27), develop OCR/HTR models for handwritten Japanese documents including kuzushiji (2027–29), and build semantic search and RAG infrastructure for Japanese texts (2027–30), culminating in a publicly available database application. The long-term collaboration with our international partners centers on smart retrieval systems for secondary literature and on language resources for Japanese translations of Buddhist literature, including automatic word segmentation and named entity tagging based on existing indices. Throughout the project, we will hold public workshops and symposia, training students and early-career researchers from across Japan in AI-driven analysis of classical texts, and present our results annually at international venues.
Building an AI Research Infrastructure for Japanese Documents: Machine Translation, Handwritten Text Recognition, and Semantic Search for Integrated Japanese Studies
Overview
Researchers
Sebastian NEHRDICH
Principal Investigator
Assistant Professor, Center for Integrated Japanese Studies, Tohoku University
Field of research: Computational Philology, Natural Language Processing
Satoshi KATO
Co-Investigator
Vice Director, Center for Integrated Japanese Studies / Professor, Tohoku University Archives, Tohoku University
Field of research: Modern and Contemporary Japanese History, Archival Science
Collaborators
・Ryuta KIKUYA
(Associate Professor, Department of Esoteric Buddhism, Faculty of Letters, Koyasan University / Center for Integrated Japanese Studies, Tohoku University)
・Oliver HELLWIG
(Researcher, Center for the Interdisciplinary Study of Language Evolution, University of Zurich)
・Sven SELLMER
(Professor, Institute of Oriental Studies, Adam Mickiewicz University)
・Jonathan SILK
(Professor, Leiden University Institute for Area Studies, Leiden University)
Project Summary Slide
Summary slide of the research overview. (Click to view larger image)