.png){kind=icon}
.png){kind=icon}
Return
Multidimensional contribution quantification strategy and XGBoost algorithm-based individual assessment method for cooperative learning
.png){kind=icon}
PRE
AI
DOI：10.1080/14703297.2025.2604183.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
For PBL practice courses teaching in college, a good collaborative learning assessment method is the key to effectively assess course outcomes. Using traditional assessment methods, for example, members share the same score and peer assessment, 'free riding' and 'subjective factors' is inevitable. This paper proposes a individual assessment method for collaborative learning based on a multidimensional contribution quantification strategy (MCQS) and XGBoost algorithm, and applies it to PBL. MCQS means teachers can customise the division perspective of project subtasks and set the weights to objectively calculating individual contribution. It is independent of the course major and it has universality. Then, a high precision and fast XGBoost regression model is trained to predict individual scores. The proposed method makes PBL assessment more accurate and objective, improves students' participation in collaboration and learning enthusiasm, and also helps students to understand assessment criteria and learning objectives more clearly.
Keywords:
Collaborative learning
PBL teaching
XGBoost
learning assessment
Journal
I
IF:
4.9
Papers:
117
Citations:
2.8K