Downes, Thomas
(2024)
Automated Grading Using Generative AI.
Master's Thesis, University of Pittsburgh.
(Unpublished)
Abstract
Recent innovations in AI have enabled a wide array of new applications to improve human
lives and productivity. These innovations enable the automation of mundane, and repetitive
tasks to an extent never before seen. In this work we leverage this technology to create a
system which can automate the time-consuming task of grading student homework submis-
sions, while also maintaining a level of quality that students expect. To aid instruction, the
system also provides a summary of common mistakes made by students and for students it
provides individualized written feedback intended to improve learning outcomes, all while
reducing the grading workload for instructors.
In addition to developing the system, we used real student data to evaluate the efficacy
and acceptance of the system. This began with using the system to grade student sub-
missions from archived semesters, and comparing the grades assigned by the TA and our
system. Next we had students in current courses engage with the system with their current
assignments. Students were provided with written feedback shortly after submitting their
homework assignment, and asked to rate how useful they found the feedback. This feedback
was chosen randomly from either the TA or the automatic grading system. Our results
show that the system is capable of automatically grading student submissions with a quality
comparable to a human grader, and provides feedback that students found helpful.
iv
Share
| Citation/Export: |
|
| Social Networking: |
|
Details
| Item Type: |
University of Pittsburgh ETD
|
| Status: |
Unpublished |
| Creators/Authors: |
|
| ETD Committee: |
|
| Date: |
9 May 2024 |
| Date Type: |
Publication |
| Defense Date: |
12 April 2024 |
| Approval Date: |
9 May 2024 |
| Submission Date: |
22 April 2024 |
| Access Restriction: |
No restriction; Release the ETD for access worldwide immediately. |
| Number of Pages: |
34 |
| Institution: |
University of Pittsburgh |
| Schools and Programs: |
School of Computing and Information > Computer Science |
| Degree: |
MS - Master of Science |
| Thesis Type: |
Master's Thesis |
| Refereed: |
Yes |
| Uncontrolled Keywords: |
AI, Automation, Grading, NLP, LLM |
| Date Deposited: |
09 May 2024 16:10 |
| Last Modified: |
09 May 2024 16:10 |
| URI: |
http://d-scholarship.pitt.edu/id/eprint/46229 |
Metrics
Monthly Views for the past 3 years
Plum Analytics
Actions (login required)
 |
View Item |
![[feed]](/images/feed-icon-32x32.png){kind=icon}