COMPUTER-BASED SYSTEM DESIGN FOR WRITING ASSESSMENT IN INDONESIAN LANGUAGE Sri Kusuma Winahyu (a*), Endry Boeriswati (b), Fathiaty Murtadho (b)
a) Universitas Negeri Jakarta
Rawamangun, Jakarta, Indonesia
*srikusumawinahyu_9906917033[at]mhs.unj.ac.id
b) Universitas Negeri Jakarta
Rawamangun, Jakarta, Indonesia
Abstract
This paper describes the design of a computer-based writing assessment system in Indonesian language, especially assessment of opinion article writing. The system designed for assessment is in the form of a computer based test (CBT) system with machine rater and human rater. The computational process is carried out by applying preprocessing steps in natural language processing (NLP). This study is based on a needs analysis for Indonesian language teachers regarding the assessment of writing opinion articles by grade XII students so that the researcher plans an assessment instrument design. In this design, opinion articles as material to be judged are taken from magazines and the role of the human rater is carried out by the researcher. This assessment design is considered important to be initiated considering that so far technology has become an absolute necessity in the teaching and learning system. The benefits of this system when associated with learning Indonesian language are in terms of efficiency and objectivity in teacher assessments of grade XII in accordance with the Basic Competencies in the 2013 Curriculum to be able to produce opinion articles. This research is also relevant to conditions today, when the Covid-19 virus outbreak occurred, which then had an impact on conditions that ^forced^ teachers and students to carry out the teaching and learning process online.
Keywords: CBT, NLP, writing assessment, opinion articles, machine rater, and human rater.
