Unsupervised Model for Detecting Plagiarism in Internet-based Handwritten Arabic Documents

Abstract

Due to the rapid increase of internet-based data, there is urgent need for a robust intelligent documents security mechanism. Although there are many attempts to build a plagiarism detection system in natural language documents, the unlimited variation and different writing styles of each character in Arabic documents make building such systems challenging. Based on its position in a word, the same Arabic letter can be written three different ways, which makes the handwritten character recognition a cumbersome process. This article proposes an intelligent unsupervised model to detect plagiarism in these documents called ASTAP. First, a handwritten Arabic character recognition system is proposed using the Grey Wolf Optimization (GWO) algorithm. Then, a modified Abstract Syntax Tree (AST) is used to match the contents of the Arabic documents to detect any similarity. Compared to the state-of-the-art methods, ASTAP improves the effectiveness of the plagiarism detection in terms of the matched similarity ratio, the precision ratio, and the processing time.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 40 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.