Data Mining Analysis for Assessing Students Proficiency in Scientific Writing

Abstract

A good understanding of the material and clear writing are important for success in academic and professional careers. However, not all students are equally skilled at writing scientific articles. This research aims to classify the levels of student understanding in writing scientific articles. This study classifies college students' understanding of scientific writing across four universities in Palembang with a sample of 108 students selected through random sampling. Data were collected via questionnaires, and the quantitative method used data mining with the C4.5 algorithm. Testing with RapidMiner software yielded a model accuracy of 74.58%. The study found that the C4.5 algorithm's accuracy in classifying students’ understanding of scientific writing falls into the Fair category, meaning the model treats all individuals or groups equally. The findings of this research should be a particular concern for higher education institutions to support and assist students in better understanding how to write scientific articles.