A Visual Content-Based Approach for Automatic Evaluation of Student Assignment Reports

Huang, Jiajun and Yang, Shuo and Xu, Chang (2019) A Visual Content-Based Approach for Automatic Evaluation of Student Assignment Reports. Australasian Association for Engineering Education 2019, Brisbane, Australia , 9-11 December, 2019.

[img] PDF (Conference handbook) - Supplemental Material
Restricted to Registered users only


Abstract or Summary

The last few years have witnessed a surge of interest in artificial intelligence (AI). There is a significant increase in the number of students who major in or take courses related to AI. This inevitably brings great pressure to unit coordinators. Especially when the deadline of assignment approaches, dozens even hundreds of assignment reports will flood in almost simultaneously. The sharp rising of workload will cause a tremendous burden to the lecturers and tutors who are responsible for marking the reports. As a result, a preliminary evaluation or a rough assessment of the reports before marking is needed to propose some referential suggestions and boost the marking process. Another severe challenge in marking a large volume of assignment reports is the marking consistency issue which has been previously discussed by Blok (1985). Due to the diversity of report markers, there is no guarantee of a consistent assessment of all the assignment reports, even if a common evaluation criterion is provided. This inconsistency problem is more serious for the novice markers. With relatively less experience, they are prone to make a biased judgment according to their subjective viewpoints. Thus, for a consistent and fair review of the student assignment reports, a more objective approach is urgent to assist the report marking. “Whoever started the trouble should end it.” The extra burden caused by the fervour of AI should be released by AI. Because of the powerful learning and generalization ability, machine learning (ML) and deep learning (DL) models have been applied to a wide range of applications including those in the education sector. Currently, the examples of intelligent education include but not limited to Intelligent Tutoring Systems (ITS) (Chaudhri et al. 2013), virtual facilitators and learning environments (Swartout et al. 2013) and so on. However, as more time-consuming and tedious job.

Item Type:Paper presented at a conference, workshop or other event, and published in the proceedings
Keywords that describe the item:Artificial intelligence, AI,
Subjects:L Education > L Education (General)
T Technology > TA Engineering (General). Civil engineering (General)
Divisions:Schools > Centre for Engineering and Industrial Design
ID Code:7018
Deposited By:
Deposited On:18 Dec 2019 20:47
Last Modified:18 Dec 2019 20:47

Repository Staff Only: item control page