Classification of MTI Student Thesis Documents at Bina Darma University Palembang Using Naïve Bayes

Authors

  • Dhea Noranita Putri 1Jurusan Magister Teknik Informatika, Universitas Bina Darma, Palembang, Indonesia
  • Tri Basuki Kurniawan Jurusan Magister Teknik Informatika, Universitas Bina Darma, Palembang, Indonesia
  • Edi Surya Negara Jurusan Magister Teknik Informatika, Universitas Bina Darma, Palembang, Indonesia
  • Yesi Novaria Kunang Jurusan Magister Teknik Informatika, Universitas Bina Darma, Palembang, Indonesia
  • Misinem Vocational Program, Universitas Bina Darma, Palembang, Indonesia

Keywords:

Data Mining, Classification, Naïve Bayes, Text Mining

Abstract

One of the resources that students might use as a guide when conducting research is the university library. A research thesis written by former students serves as reference material. Students must arrange the thesis documents following the concept or topic of their research because they are typically organized by faculty and department. Researchers, therefore, attempt to classify student thesis documents according to themes or subjects so that students can be more precise in their search for references to themes or topics that relate to the research they will do. The title, abstract, and important keta from the thesis document will be used as the study's data, which will then be classified using the best classification technique, the Naive Bayes Classification (NBC) approach. The learning stage and the testing stage are the two steps used in the naive Bayes classifier method's classification process. After establishing the Category and the quantity of data learning documents, probability calculations were then carried out for each category.

Published

2022-12-15