Guo S1, Sun C1
1Sydney Informatics Hub, Core Research Facilities, The University of Sydney, Camperdown, Australia
Biography:
Senhui (Alex) Guo is a Data Science Software Engineer in the Sydney Informatics Hub at the University of Sydney, where he leads development of LDaCA Wordflow — a cross-platform text analytics application built for the Australian Text Analytics Platform (ATAP), part of the Language Data Commons of Australia. His work focuses on research software engineering: lowering the barrier to computational text analysis without sacrificing rigour. Senhui holds a PhD in Electrical and Information Engineering from the University of Sydney and a Bachelors in Mathematics and Applied Mathematics from the University of Science and Technology of China.
Abstract:
For the past few years, the team has developed text analytics tools for humanities and social science researchers as a collection of Jupyter notebooks, under the LDaCA text analytics program (formerly the Australian Text Analytics Platform, ATAP). The notebooks did their job, and taught us their limits: environments drifted, each tool was maintained in isolation, the user experience fragmented across tools, and researchers who don't code hit a hard ceiling. In 2025 we took a different approach, rebuilding and consolidating the collection into LDaCA Wordflow, a single open-source, browser-first platform.
The rebuild brought together a React front end, a FastAPI back end, and a Rust-powered Polars data engine that keeps million-row corpora interactive on a standard office computer; a visual workspace in which the corpus and every analysis result live as connected data blocks, so most results can feed the next tool for follow-up analysis, and the workspace itself becomes both a provenance record and a shareable digital asset; language-appropriate tokenisation across English and other European languages, Chinese, Japanese, Korean, and more; and built-in features that make point-and-click analyses reproducible and shareable, both between collaborators and across deployment platforms.
Wordflow has been in active development for a year, with several public releases, each archived with a digital object identifier (DOI). From a single codebase, it is delivered to cloud, web, command-line, and signed desktop applications, and has been tested by researchers in hands-on workshops and real research projects. We share what worked for us, the payoffs of this substantial investment, and how the rebuild lowers the barrier for the researchers we support to take up computational text analysis.
