Web-based automatic assessment for comprehension questions in PSLE Chinese
The Primary School Leaving Examination (PSLE) is an annual national examination in Singapore administered by the Ministry of Education and taken by all students near the end of their sixth year in primary school before moving on to secondary school. The PSLE Chinese Language (CL) examinations are pi...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project |
| Language: | English |
| Published: |
2015
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/10356/62900 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Nanyang Technological University |
| Language: | English |
| Summary: | The Primary School Leaving Examination (PSLE) is an annual national examination in Singapore administered by the Ministry of Education and taken by all students near the end of their sixth year in primary school before moving on to secondary school. The PSLE Chinese Language (CL) examinations are pitched at different levels for students taking Standard CL and Higher CL, to assess students’ different levels of ability and attainment. In order to do well during examination, it is important to have enough practice. However, it is a time consuming and tedious process for teachers to mark students’ answers. Therefore, it is highly desirable to be able to assess students’ answer automatically. Various tools for automatic grading English Language are available. However, no known system can reliably perform similar grading for the Chinese Language. This project focuses on Comprehension questions in PSLE Standard CL. It aims to develop a web-based system for automatic assessment of short answers with instant feedback. There are mainly two grading criteria to assess student answers: (1) semantic relevance and (2) syntactic correctness. In this project, we focus more on semantic analysis for short answers assessment. An ensemble approach that combines Latent Semantic Analysis, n-grams and BLEU is proposed for semantic assessment. In this report, various techniques for Chinese word segmentation and automatic short answer grading will be reviewed. The design of the semantic module and syntactic module will be presented. Then, the presented system will be evaluated by comparing its scores with the score given by a human grader. Finally, the directions for future research work are given. |
|---|