Using a data-driven statistical modelling and analysis approach to model individual learning progress

Project: Research


This industry collaborative research proejct has the following objectives
1. Develop best practices or guidelines that would be needed to assist teachers and students to make use of the information provided by learning systems in terms of (i) data analysis and visulisation, and (ii) corresponding actions or interventions within the classroom.
2. Utilise existing functionality within the learning system to assist in developing ML/DL-based approaches to make use of learning progress data to inform classification, prediction, and recommendation processes within Virtuoso. In particular, utilise ML, Sentiment/Emoitional text analysis along with Topic Modelling etc. to provide improved and actionable insights based on student feedback

The research targets a data driven methodology to address both of these objectives.The research explores exisiting literature to determine the best practices to help the end user (teachers and students) with improved analytics and data visualisation. The methodology will consist of reviewing existing practices to improve the system and validation based on the user feedback. The methodology will be reviewed based on the requirement but in general will consist of statistical validation of the data driven process.

Layman's description

This is an industry collaborative research project with Cinglevue International is working to explore the applications of data-driven analysis in modelling individual learning progress. This collaborative research project involves analysing student feedback to determine the extent of engagement and understanding via topic modelling and sentiment analysis as well as establishing guidelines to assist educators to make effective use of this information as well as other forms of teaching and learning data.
Effective start/end date1/02/1831/12/18