Open Access

ARTICLE

A Reliable NLP Scheme for English Text Watermarking Based on Contents Interrelationship

by Fahd N. Al-Wesabi^1,2,*, Saleh Alzahrani³, Fuad Alyarimi³, Mohammed Abdul³, Nadhem Nemri³, Mohammed M. Almazah⁴

1 Department of Computer Science, King Khalid University, Muhayel Aseer, Kingdom of Saudi Arabia
2 Faculty of Computer and IT, Sana’a University, Sana’a, Yemen
3 Department of Information Systems, King Khalid University, Muhayel Aseer, Kingdom of Saudi Arabia
4 Department of Mathematics and computer in Ibb University, Yemen and Department of Mathematics, King Khalid University, Muhayel Aseer, Kingdom of Saudi Arabia

* Corresponding Author: Fahd N. Al-Wesabi. Email:

Computer Systems Science and Engineering 2021, 37(3), 297-311. https://doi.org/10.32604/csse.2021.015915

Received 01 December 2020; Accepted 06 January 2021; Issue published 08 March 2021

Abstract

In this paper, a combined approach CAZWNLP (a combined approach of zero-watermarking and natural language processing) has been developed for the tampering detection of English text exchanged through the Internet. The third gram of alphanumeric of the Markov model has been used with text-watermarking technologies to improve the performance and accuracy of tampering detection issues which are limited by the existing works reviewed in the literature of this study. The third-grade level of the Markov model has been used in this method as natural language processing technology to analyze an English text and extract the textual characteristics of the given contexts. Moreover, the extracted features have been utilized as watermark information and then validated with the attacked English text to detect any suspected tampering occurred on it. The embedding mechanism of CAZWNLP method will be achieved logically without effects or modifying the original text document to embed a watermark key. CAZWNLP has been implemented using VS code IDE with PHP. The experimental and simulation results using standard datasets of varying lengths show that the proposed approach can obtain high robustness and better detection accuracy of tampering common random insertion, reorder, and deletion attacks, e.g., Comparison results with baseline approaches also show the advantages of the proposed approach.

---

Keywords

---