Piloting the FeedbackFruits A.I. automated feedback tool
Deakin has partnered with FeedbackFruits to develop an artificial intelligence (A.I.) powered tool for automated feedback on academic writing for our students. We spoke to Laura Tubino, Lecturer of Curriculum Design (SEBE) in the Deakin Learning Futures team about how the tool has been piloted and what are the next steps in its design and implementation.
How does the FeedbackFruits A.I. automated feedback tool work?
The tool harnesses the power of A.I. and machine learning to provide instant formative feedback on multiple elements in academic writing texts. It covers elements such as structure and content, academic language, citations and references, tables and figures. Elements are being added continuously based on user feedback.
What is the benefit for students?
Students can upload their tasks to the tool and get immediate feedback on grammar, structure, and referencing before submitting. This tool is usually included as an optional step to help students improve their submission, with students being able to use it for formative feedback as many times as they need.
What stage is the project in?
Deakin has partnered with FeedbackFruits to design the tool. FeedbackFruits prepared the initial tool which Deakin piloted in Trimester 2 2021 in units from the Faculty of Sciences and Built Environments (SEBE).
In Trimester 3 2021 and Trimester 1 2022 we launched a more extensive pilot. In T1 2022 the tool is being used in 15 units across faculties and is available to 3800 students.
How have students and staff responded to using the tool?
We’ve found that 15% of students involved in the pilot have used the tool, which is great for a non-compulsory tool. The students that have used the tool more than once tend to use it again on the same day, which means students are finding the tool useful. They are uploading a task, getting feedback, making corrections, and then uploading the next draft to the tool to recheck their writing before submission.
We’ve also had a positive response from academics involved in the pilot, who have indicated that they would like to continue using the tool in the future and in other units they teach outside of the pilot.
What are the next steps?
The next stage is added functionality to give students more control over how they use the tool. We want students to be able to set up the tool and choose what elements of feedback they are looking for in their document. This will improve students’ academic writing and support them to become life-long learners by developing their feedback literacy and evaluative judgment.
We are codesigning the tool with students and running a series of prototype testing sessions where students across all the faculties can provide their feedback on the design of the student facing version of the tool.
You can hear more about the FeedbackFruits A.I. automated feedback tool in episode 40 of the Tales of Teaching Online podcast.