Scholarlink

Threats through anonymous letters continue to be reported in Indonesia. Threatening letters are being sent via SMS, e-mail, and private messaging apps such as WhatsApp. Authorship analysis is commonly used to determine the author of an anonymous text. The study uses N-gram tracing to determine the author of an anonymous letter by analyzing three sets of texts from three unique authors. The data was analyzed by determining, tracing, and computing N-grams in each set of texts being compared, both at the character and word levels. Statistical tests were also performed during the data analysis stage, utilizing the similarity comparison method and the Jaccard Coefficient calculation to assess the accuracy of N-gram tracing in identifying authors. Character analysis of N-grams reveals that, as the smallest N-unit, characters play an essential role in authorship attribution. Data analysis at the word level reveals that, lexically, word choice is the most dominating and influential linguistic element of authorship attribution in defining the author's profile and accurately distinguishing one author from another.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

N-gram Based Authorship Analysis in Indonesian Text: Evidence Case Study in Authorship Dispute Cases

  • 章节

  • 出版模式：

    开放获取

  • 发表日期：

    2025年

查看全文
  • Devi Ambarwati Puspitasari,
  • Adi Sutrisno,
  • Hanif Fakhrurroja

摘要

Threats through anonymous letters continue to be reported in Indonesia. Threatening letters are being sent via SMS, e-mail, and private messaging apps such as WhatsApp. Authorship analysis is commonly used to determine the author of an anonymous text. The study uses N-gram tracing to determine the author of an anonymous letter by analyzing three sets of texts from three unique authors. The data was analyzed by determining, tracing, and computing N-grams in each set of texts being compared, both at the character and word levels. Statistical tests were also performed during the data analysis stage, utilizing the similarity comparison method and the Jaccard Coefficient calculation to assess the accuracy of N-gram tracing in identifying authors. Character analysis of N-grams reveals that, as the smallest N-unit, characters play an essential role in authorship attribution. Data analysis at the word level reveals that, lexically, word choice is the most dominating and influential linguistic element of authorship attribution in defining the author's profile and accurately distinguishing one author from another.