Analisis Pengaruh Teks Preprocessing Terhadap Deteksi Plagiarisme Pada Dokumen Tugas Akhir

Abstract

Final Project Report at a university has the potential for plagiarism. To detect possible plagiarism, String Similarity can be used. Text preprocessing is needed to process words which can make String Similarity results inaccurate. The value of the distribution of the results of the similarity that is getting higher shows the level of accuracy is also getting higher. Reports that contain many words can make it difficult to find plagiarism recommendations. In this study, we try to divide the report into each chapter to provide more detailed recommendation material. By using text preprocessing and comparison methods in the same chapter, can determine the characteristics of each chapter. The discovery of the characteristics of each chapter can be used as plagiarism recommendation material in more detail than a full text report. The experiment was a comparison of the results of cosine similarity between the same chapters and full text, then combined with preprocessing stopword removal and stemming. The experimental results show that the use of preprocessing stopword removal and stemming can produce the highest distribution value and the similarity ratio in each chapter can show its characteristics. Words that represent the characteristics of a chapter can potentially become a stopword.

Analisis Opini Publik Terhadap Undang-Undang KUHP Tahun 2022 Menggunakan Algoritma Naïve Bayes Classifier

“…Pada tahapan Stopword, menghapus kata yang dianggap tidak memberikan dampak terhadap informasi yang ada pada suatu data tweet. Pembersihan data Stopwords pada penelitian ini menggunakan library python pySastrawi [19].…”

N-gram and Kernel Performance Using Support Vector Machine Algorithm for Fake News Detection System

“…Stop word usually have little or no meaning, for example, the words "which," "in," "to," "from". The main function of stop word removal is so that stop words do not affect the results of the next process [14]. The final step of preprocessing is stemming.…”

Analisis Sentimen Terhadap Isu Penundaan Pemilu di Twitter Menggunakan Naive Bayes Clasifier

“…Pembersihan data dilakukan untuk membersihkan url, mention, hashtag, stopwords, emoticon, tanda baca, dan angka, juga mengecilkan huruf. Pembersihan data stopwords pada penelitian ini menggunakan library python pySastrawi [12]…”