Non-Relevant Document Reduction in Anti-Plagiarism Using Asymmetric Similarity and AVL Tree Index

Non-Relevant Document Reduction in Anti-Plagiarism Using Asymmetric Similarity and AVL Tree Index

Author		: ADEVA OKTOVERI; AGUNG TOTO WIBOWO; ARI MOESRIAMI BARMAWI
Published on	: The 5th International Conference on Intelligent and Advance Systems (Universiti Teknologi Petronas - Kuala Lumpur Malaysia)

Abstract

Anti-plagiarism applications have been developed using various approaches. Many methods compare one document to the others, regardless it is relevant or not. This paper proposed a method to reduce non-relevant document (documents that have no similar topic with query document) by using asymmetric similarity. Whole documents are collected in one corpus. For each document, it will be preprocessed using winnowing algorithm. The feature from winnowing is then indexed using AVL Tree algorithm to fasten the documents comparing process. The result shows that reducing non-relevant document shorten the time processing almost 10 times compared to non-reduced process. Meanwhile, both processes show the same accuracy to give suspected documents that is 89.78%.