AI tools for reading, searching and analysing historical Malay documents in Jawi script.
AI tools for historical documents in Malay (in the Jawi Script). This project includes a range of tools for OCR, transliteration, classification, information extraction and semantic search. We aim to enable the systematic analysis of Jawi print media published in Singapore and its neighbors.
Segmenting recordings of performances into meaningful actions, with Wayang Kulit as a case study.
A complete pipeline for collecting, annotating, and modeling temporal action segmentation data—with a special focus on cultural analytics. Our case study centers on Wayang Kulit (Indonesian shadow puppetry), showcasing how machine learning can be applied to analyze performance-based materials.
Using LLMs to simulate how different communities might respond to heritage policies.
Using LLMs to simulate the range of reactions different communities might have to heritage-related policies, in order support more open, informed public dialogue about shared heritage. This is part of a larger programme which explores LLM-based social simulation for policy design.
