The JK System to Detect Plagiarism

• Autores: Sabri, Khair Eddin M.; Al-Jaafer, Jubair J.
• Año de publicación: 2006
• Idioma: inglés
• Tipo de recurso: artículo
• Materia: Ciencias Informáticas; SOFTWARE ENGINEERING; Entropía; Plagio; Restructuring, reverse engineering, and reengineering
• Institución: Universidad Nacional de La Plata

In this research a system, referred to as Jubair-Khaireddin (JK), has been developed to assess the degree of similarity between two programs even though they look superficially dissimilar. The JK system has the capability to detect deliberate attempts of plagiarism. Reverse engineering technique is used to bring each suspected program back to its initial specification stage. This operation enables us to extract the structure of the program which is an important factor in detecting plagiarism. This can be achieved through the extraction of the Static Execution Tree (SET) for each program. The SET is then transformed into Terminating Binary Sequence (TBS). The TBSs generated from the tested programs are compared in order to get similar branches. Reengineering technique is then applied on these similar branches in order to compute its entropy (information content). The entropy is computed to prove or disprove the existence of similarities between programs. The JK system has been tested on different Java programs with different modifications, and proved successful in detecting almost all cases including those of partially plagiarised programs.
Facultad de Informática