TY - GEN
T1 - Automated Grading of Discussion Posts in Online Courses
AU - Dumoulin, Gabriel
AU - Sakeef, Nazmus
AU - Dewan, M. Ali Akber
AU - Wen, Dunwei
N1 - Publisher Copyright:
© 2025 IEEE.
PY - 2025
Y1 - 2025
N2 - Discussion forums and online courses have been growing in popularity due to recent advancements in education and technology. However, the means of effectively managing these tools have not been fully developed. This paper proposes a method for automatic grading of students' posts in course discussion forums as an alternative approach to manual evaluation. The method employs a machine learning based approach to analyze students' posts on three key aspects: engagement, post relevance, and writing quality. The aspects are analyzed individually to enhance the model's precision and combined to provide instructors with a flexible grading system. The auto grading system is developed using GloVe embeddings, Sentence-BERT, and Term Frequency-Inverse Document Frequency (TF-IDF) along with other textual features and machine learning model. This system is developed with the intend to reduce instructors' workload, provide students with self-assessment opportunities, and enhance their comprehension and engagement in courses.
AB - Discussion forums and online courses have been growing in popularity due to recent advancements in education and technology. However, the means of effectively managing these tools have not been fully developed. This paper proposes a method for automatic grading of students' posts in course discussion forums as an alternative approach to manual evaluation. The method employs a machine learning based approach to analyze students' posts on three key aspects: engagement, post relevance, and writing quality. The aspects are analyzed individually to enhance the model's precision and combined to provide instructors with a flexible grading system. The auto grading system is developed using GloVe embeddings, Sentence-BERT, and Term Frequency-Inverse Document Frequency (TF-IDF) along with other textual features and machine learning model. This system is developed with the intend to reduce instructors' workload, provide students with self-assessment opportunities, and enhance their comprehension and engagement in courses.
KW - automated grading
KW - forum posts analysis
KW - machine learning
KW - MOOC
KW - natural language processing
KW - online learning environment
UR - https://www.scopus.com/pages/publications/105035828692
U2 - 10.1109/SWC65939.2025.00042
DO - 10.1109/SWC65939.2025.00042
M3 - Published Conference contribution
AN - SCOPUS:105035828692
T3 - Proceedings - 2025 IEEE Smart World Congress, SWC 2025, 2025 IEEE Ubiquitous Intelligence and Computing, Autonomous and Trusted Computing, Digital Twin, Metaverse, Scalable Computing and Communications
SP - 90
EP - 95
BT - Proceedings - 2025 IEEE Smart World Congress, SWC 2025, 2025 IEEE Ubiquitous Intelligence and Computing, Autonomous and Trusted Computing, Digital Twin, Metaverse, Scalable Computing and Communications
T2 - 2025 IEEE Smart World Congress, SWC 2025
Y2 - 18 August 2025 through 22 August 2025
ER -