Home / News / Open Source Writing Analytics

Open Source Writing Analytics

Posted on


CIC has released its Academic Writing Analytics (AWA) infrastructure open source.

THE CHALLENGE. CIC initiated its Academic Writing Analytics (AWA) project in 2015, as it became clear through consultations across faculties that student writing was a strategically important area for UTS teaching and learning (and indeed, for most other educational institutions). The possibility of providing instant, personalised, actionable feedback to students about their drafts, 24/7, is a compelling one.

To deliver on this vision requires integrated expertise including natural language processing, linguistics, academic language pedagogy, learning design, feedback design, user experience, and cloud computing. Truly a transdisciplinary effort, which has been enormously stimulating.

PROGRESS. This is not solved! But we’ve made good progress. To date, we have worked on critical, argumentative, analytical writing of the sort typically found in literature reviews, persuasive essays and research articles, as well as reflective writing, in which learners make sense of their workplace experiences, try to integrate this with their academic understanding, and share their own uncertainties, emotions and sense of personal challenge/growth. The goal is to more effectively teach the building blocks of good academic writing. Naturally once these have been mastered, skilled writers can break the rules in creative ways — but far too many students have never had the features of good moves explained well to them, and most educators are subject matter experts, not skilled writing coaches.

The team has developed a series of prototype web applications, culminating in AcaWriter, powered by a modular Text Analytics Pipeline. This has been piloted in a range of contexts across the university, in close partnership with educators. We’ve been sharing what we’re learning via peer reviewed research publications and hands-on workshops in institutional, national and international fora, and we continue to collaborate to mutual benefit with other writing analytics research groups doing complementary work.

ACADEMIA/INDUSTRY PARTNERSHIP: This progress has been made as a direct result of an extremely fruitful research collaboration with Naver Labs Europe (formerly Xerox Research Centre Europe). The Athanor tool and the rhetorical analysis resources were implemented in collaboration with Claude Roux and Ágnes Sándor, to whom we are indebted.

OPEN SOURCE. As envisioned at the dawn of the discipline of Learning Analytics as a community (see this 2011 Open Learning Analytics white paper from founding members of the Society for Learning Analytics Research), we can accelerate learning analytics innovation and algorithmic accountability through open source infrastructure. We are therefore delighted to announce the open source release of the AWA infrastructure, as part of the new Higher Education Text Analytics (HETA) project, funded by the Australian Technology Network of universities.


Integrated educational resources: the HETA resources page introduces how we embed a given version of AcaWriter in a learning context, with associated Writing Activities with Writing Analytics (WAWA) resources (Creative Commons CC-BY-SA license).

Code: the HETA technology page introduces the components of the technical infrastructure, with links to the live demo, GitHub repositories (Apache 2.0 license): Text Analytics Pipeline (TAP), Athanor (rhetorical parser), Jupyter notebooks in Python. Quick links:

JOIN THE COMMUNITY. We encourage you to set up your own instances, and let the technical team know how it’s going (via Github). Start building and sharing your own extensions, whether code, integrating new text analysis services into TAP, devising new AcaWriter Genre profiles (to identify features in a given text type and design associated feedback annotations and messages), or Learning Designs. Join the HETA Writing Analytics google group to discuss effective uses and evolution of the tools.

There is much more to do. We hope this gives you a springboard into automated writing feedback, and we invite you to work with us on providing better feedback on writing, to far more learners than used to be possible.

One thought on “Open Source Writing Analytics

Comments are closed.