The judgment of document similarities using various orthogonal transformations
Atraru Matsuzawa, Masakazu Higuchi, Jonah Gamba, Shuji Kawasaki, Hitomi Murakami · International Conference on Circuits · 2011
The objective of this work is to propose a new method to evaluate similarity of documents using Orthogonal transformations and examine its performance experimentally expecting its application to plagiarism detection. Specifically, it is carried out to each of Japanese characters corresponds a Shift-JIS code of two-byte decimal numbers in a computer. The numbers are transformed by the Fourier, Cosine, Haar Wavelet and Hadamard system. As a measure of the similarity of two documents, the correlation of them is used. As a result of the experiment, it turns out that Fourier transformation is effective for certain similar documents presumed to be produced by plagiarism.