Penerapan Algoritma Knuth Morris Pratt Untuk Mendeteksi Tingkat Plagiarisme Pada Tugas Mahasiswa

The rapid development of information technology has made accessing information easier but has also increased the likelihood of plagiarism among students. Therefore, detecting the level of plagiarism in student assignments is crucial to maintaining academic integrity. This study employs the Knuth-Morris-Pratt Algorithm supported by the Jaccard Similarity method to detect similarities between student essay assignments in Indonesian, stored in Word and PDF formats without images or tables. The system is developed using the Python programming language and the Flask framework, complemented by optimized data preprocessing to enhance detection efficiency and accuracy. The testing results indicate that the Knuth-Morris-Pratt Algorithm can detect plagiarism levels with an average accuracy of 98.66% and classify plagiarism detection results into appropriate categories, such as Minor Plagiarism, Moderate Plagiarism, and Severe Plagiarism.