Discrepancies Detection in Arabic and English Documents

In the paper, there are analyzed and compared results of usable methods for discrepancies detection based on character n-gram profiles (the set of character n-gram normalized frequencies ofa text) for English and Arabic documents. English and Arabic texts were analyzed from many statistical characteristics point ofview. We covered some statistical differences between both languages and we applied some heuristics for measurements oftext parts dissimilarities. The results for each text can call for an attention to the text (or not) if the text parts were written by thesame author. We evaluate some Arabic and Englishdocuments and show its parts they contain discrepancies and they need some following analysis for plagiarism detection. The analysis depends on selected parameters prepared in experiments.

کلیدواژه ها:

n-grams ، stylistic measure ، plagiarism ، authorship