TurnItIn to identify third-party authorship with machine learning

University anti-plagiarism software company TurnItIn will bring out a new product that uses machine learning to tackle the prevalence of students engaging a third party to complete their assessments, sometimes referred to as contract cheating.

The new software, called ‘Author Investigation’, will identify instances of contract cheating by building a profile of a student’s writing style and using machine learning algorithms and ‘forensic linguistic best practices’ to detect major differences in a student’s style.

According to preliminary results from a recent survey on contract cheating at Australian universities, 68% of teaching staff suspect that their students may have submitted work that they did not author.

The software will be made available in the second half of 2018, with a number of Australian universities advising on the current development process. They include Deakin University, Griffith University, the University of New South Wales, the University of Queensland and the University of Wollongong.

‘A new industry is servicing a demand for assessment work to be completed to order for students so it goes unnoticed by existing detection tools. Something that enables institutions to identify when this is happening in a way that is both efficient and effective is going to be a big step forward,’ said Cath Ellis, associate dean (education) in the Faculty of Arts and Social Sciences at UNSW.

For more information, click here.

Category: Library news