Hand-ins compared with each other, with earlier hand-ins for the same subject and with the subject’s own material, inside the institution. It is part of the assignments workflow in the document locker, not a standalone plagiarism service, and it does not search the internet or published work.
Each hand-in is compared with the other hand-ins for the same assignment, with earlier years’ hand-ins and assignment briefs for the same subject, and with the subject’s study material in the document locker. Everything stays inside the institution; nothing is sent to an outside service.
It does not search websites, journals, books or commercial databases, so text copied from a source that is not in the locker will not be found. That needs a licensed external service, and none is connected as standard. The report says what was compared, so a low figure is not mistaken for proof of originality.
Which of these are included in a deployment is set out in the proposal for that institution.
Faculty tick similarity checking on an assignment and run the check from its submissions list. Each hand-in is read as it arrives, so later hand-ins are compared with earlier ones.
PDF files with a text layer, Word .docx, plain text and web pages. A scanned PDF with no text layer is reported as such rather than scored.
Each text is reduced to fingerprints of overlapping six-word sequences, using a technique known as winnowing, and a matched run must be at least ten words long to count. Words in Indian scripts such as Devanagari are read as words, not discarded.
Passages in quotation marks, a closing references list and text taken from the assignment’s own brief. A phrase found across much of the subject’s work is treated as the subject’s common language.
A percentage for each hand-in, with its text shown and matched passages highlighted in a colour per source, each set beside the source passage it matches.
A marker sees another student’s words only for classes they teach. A match with a class they do not teach is shown without the name or the text.
Optionally, a student sees their own percentage and highlighted passages, never whose work they match, and can hand in a corrected file before the due date.
For group assignments, every member of the group sees the group’s hand-in and its report as their own.
The extracted text is covered by the institution’s retention schedule, and a student’s hand-in files, extracted text and fingerprints are removed when their personal data is erased.
An illustration of a typical sequence. Your own process is configured during implementation.
Faculty set an assignment in the document locker and switch on similarity checking.
Students hand in; each file’s text is read as it arrives.
If the student view is on, a student can see their own result and hand in a corrected file before the due date.
After the deadline, the marker runs the check from the submissions list and sees a percentage for each hand-in.
The marker opens the report for any hand-in that needs a closer look and reads each match beside its source.
The marker decides what, if anything, follows, under the institution’s own academic integrity rules.
Similarity checking is part of the document locker module of OLT SIS, which holds study material and assignments. It uses the platform’s students, subjects and faculty assignments. See the OLT SIS platform.
The institutions we have worked with, and the letters they have written, are on our clients page. In a demonstration we show this module working on demonstration data.
Course pages, study material and assignments, video, SCORM lessons, LTI tools and per-student learning progress.
One student record from registration to transcript, covering courses, attendance, assessment, grading and attainment reporting.
The AI add-on for faculty, which works inside the quizzes and document locker modules.
No. It compares a hand-in with the other hand-ins for the same assignment, with earlier hand-ins and assignment briefs for the same subject, and with the subject’s material in the document locker. Nothing is sent to an outside service, so it cannot find text copied from websites, journals or books that are not in the locker. That needs a licensed external service, and none is connected as standard.
No. A match is a reason to look, not a finding. Common phrasing, a source two students both quoted without quotation marks, or agreed group work can all match. The report shows each matched passage beside its source so that the marker can judge.
Similarity checking is part of the document locker module rather than a separately priced product. The cost depends on the modules you take, the number of students and the deployment model. Tell us those and we will give you a figure.
Tell us how your faculty set and mark written work today and the demonstration will follow that.
