An Adaptive Plagiarism Detection System Based on Semantic Concept and Hierarchical Genetic Algorithm

Abstract

Plagiarism became a considerable issue; the reason is easy access to articles on the Internet. However, many issues arise as the majority of available Plagiarism Detection (PD) tools could not identify plagiarism by structural variations and paraphrasing. For applied systems, with regards to more complicated levels, those systems fail. Genetic Algorithm (GA) is now broadly utilized in accomplishing best solution in multidimensional nonlinear problems, unfortunately, system structure must be pre-defined to be optimized. This paper introduced an improved plagiarism detection system aiming to detect cases of plagiarism by semantic similarity with Hierarchal Genetic Algorithm (HGA). HGA operates without pre-defining system structure, moreover, system structure and parameters might be optimized. For discovering plagiarism, semantic similarity depending on intelligent procedures must be applied for extracting the idea. In addition, HGA is employed in finding interrelated cohesive sentences that convey the concept. Results reveal the capability of the system to present a significant improvement over compared systems.