Development of a Plagiarisms Detecting System for Reporting Assignments based on the Style Model

2015 Fiscal Year Final Research Report

• Project/Area Number: 25330408
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Learning support system
• Research Institution: Kobe University
• Principal Investigator: Murao Hajime 神戸大学, 国際文化学研究科, 教授 (70273761)
• Project Period (FY): 2013-04-01 – 2016-03-31
• Keywords: 盗用発見 / 機械学習 / 教育支援

Outline of Final Research Achievements

In this study, we have developed a prototype system to detect plagiarisms in reporting assignments. The system firstly train a Hidden Markov Model (HMM) for each of students from "expression feature" of submitted reports, such as a number of punctuations, the location of new-lines, etc. We call the trained HMM a "style model". Then, whenever a student submits a new report, the system uses the student’s style model to validate if the report has been actually written by the student. We have tested the system in our laboratory and observed quite fair result around 70% of correct classification.

Free Research Field

社会システム科学