Subjective Answer Marking Using Keyword Extraction

Abstract

The Covid-19 pandemic gave a significant impact on educational institutions throughout Malaysia and had caused these institutions to diverse their teaching and learning method from face to face to online learning to ensure continuous learning can be implemented optimally. This situation also forced all assessments (final exams, quizzes, tests, etc.) to be done online. These assessments are marked manually either the lecturer or teacher downloads the students’ answers and prints them or marks the answers digitally using any available apps. This affects the lecturers or teachers in many aspects, especially computer related health problems due to long use of digital devices and looking at the monitor screen. Therefore, this study aims to develop a web-based system that can assists marking process especially for subjective answers using keyword extraction approach. The system is developed and utilized python and flask micro web framework. The keywords similarity being tested to compare the student’s answer to an answer scheme. The marks from the automated evaluation and manual evaluation by the lecturer were compared and the differences were calculated. The results of the automated marks are approximately as the same as manual marking with a little difference value.