Impact of different temporal lenghts on collaboration quality prediction
Python
Investigated impact of different temporal lengths on the performance of collaboration prediction model
Project overview
This project aimed at identifying optimal temporal window size to aggregate multimodal data into features that can improve generalizability of collaboration prediction models. The project analyzed audio and logs data collected from groups working to solve a given problem in collaboration.
Research Paper
The findings of this project were published in the Learning Analytics & Knowledge conference organized in Arizona, USA in 2023. You can access the paper here
⚡ Key Contributions
- Processed audio and logs data using different temporal window sizes
- Developed machine learning models for predicting collaboration using processed data
- Evaluated models at different levels of generalizability
Skills Applied
Python Programming
, Web Development
, DevOps
, Bayesian Modeling
Libraries Used
pgmpy
,pandas
,matplotlib
,numpy
,dash