Enhancing plagiarism detection methodology with the DQN algorithm on an improved differential evolution foundation

Abstract

Plagiarism detection has become increasingly crucial in real-world applications, demanding precise identification of content similarity. This paper introduces a novel plagiarism detection approach. Building upon LSTM as the foundation, it employs an enhanced DE (Differential Evolution) algorithm and reinforces learning with the DQN algorithm for sample classification and training. Throughout the training process, gradual parameter adjustments are made with the aim of improving the model's efficiency and accuracy.