Researchers based at the University of Copenhagen have demonstrated, through the use of big data analysis and artificial intelligence, that computer programs are now very effective at assessing examination papers and assignments for originality. Technology can now determine whether the student wrote the assignment or whether someone else penned it. The latest forms of technology can do this with close to 90 percent accuracy.
The researchers have constructed a ghostwriter program which utilizes a Siamese neural network. This process can distinguish the writing styles of two texts. Over time the network is trained using voluminous amounts of data to learn from representations of writing styles (in this case, 130,000 essays were examined from 10,000 students). These are the compared by the program. Siamese neural networks are also being used for recognizing handwritten checks, automatic detection of faces in camera images, and matching queries with indexed documents.
With the new software when a student submits an assignment, the network can scan the document and compares the text against previous assignments from the same student, looking at things like word length, sentence structure and how words are used. With each previous assignment, the network will provides a percentage score for writing style similarity compared with the new assignment. This produces a weighted average of the scores, which is used to indicate the similarity between the new assignment and the student’s writing style.
As to if and when the technology will be used, an ethical discussion in relation to how the technology will be applied is necessary. The inventors of the technology note that the result delivered by the program should not stand on its own, and instead function to support and substantiate a suspicion of cheating.
The program was put together by a group at the university called DIKU-DABAI (Danish Center for Big Data Analytics driven Innovation), led by Professor Stephen Alstrup. A white paper charting the development of the software has been produced, titled “Detecting Ghostwriters in High Schools.”
