Automated PSLE English situational writing practice system

• Main Author: Lim, Miranda Li Qun
• Other Authors: Hui Siu Cheung
• Format: Final Year Project
• Language: English
• Published: 2017
• Subjects: DRNTU::Engineering::Computer science and engineering
• Online Access: http://hdl.handle.net/10356/70235
• Institution: Nanyang Technological University

Even with today’s technologies, essay grading is still done manually by teachers. This process is both time-consuming and inefficient. It is also difficult to achieve a non-subjective grading of all students’ essays. Automated essay grading systems have been built over the years in an attempt to solve these problems. However, despite continuous improvements made to these systems, many were still unable to achieve full automation. One such system is the Automated Primary School Leaving Exam (PSLE) Situational Writing Practice System. Students are able to practice mock essay questions and gain content and language feedback from the system. Due to the low accuracy in detecting grammar and semantic errors, full automation has yet to be achieved for this system. This project aims to improve the system by conducting an evaluation of the grammar and semantic analysis of the system. Much of the research has been devoted to analyzing the errors made in the grammar rules and proposing solutions to solve these errors.