Deteksi keberadaan kalimat sama sebagai indikasi penjiplakan dengan algoritma hashing berbasis N-Gram

D Purwitasari, PY Kusmawan… - Jurnal Ilmiah KURSOR, 2011 - kursorjournal.org
D Purwitasari, PY Kusmawan, UL Yuhana
Jurnal Ilmiah KURSOR, 2011kursorjournal.org
Abundant cases of plagiarism committed by some intellectual people in the Indonesia’ s
education fields have became such tragedy. Due to the amount of information which is
available online are things that make copy-paste without proper citation cause plagiarism.
This paper discusses about how to detect similar sentences which is probable caused by
copy-paste. However plagiarism detection still needs further examination such as the
existing of citation or not. Winnowing algorithm is used for detecting similar sentences …
Abstract
Abundant cases of plagiarism committed by some intellectual people in the Indonesia’ s education fields have became such tragedy. Due to the amount of information which is available online are things that make copy-paste without proper citation cause plagiarism. This paper discusses about how to detect similar sentences which is probable caused by copy-paste. However plagiarism detection still needs further examination such as the existing of citation or not. Winnowing algorithm is used for detecting similar sentences between text files which is treated as a common subsequence problem. The algorithm finds document fingerprinting by changing sequence of N-grams from text into a set of hash values. If copy-paste sentences are found then both of text files must have the same document fingerprinting. Experiment has been done to observe the capability of detecting similar sentences by analyzing on value variations of n-gram, prime base b for hashing, window w, and threshold for determining plagiarism indication.
kursorjournal.org
以上显示的是最相近的搜索结果。 查看全部搜索结果