Doctoral project: Visual learning analytics to support teachers’ pedagogical work
This doctoral project features a collaboration with a number of EdTech companies that develop digital learning materials. The goal is to interpret and model the educational data with the help of visual learning analytics, and to develop a tool for teachers to support decision-making and learning trajectories and to promote pupils' study achievements.
Doctoral student Zeynab Mohseni Supervisors Italo Masiello, Rafael Messias Martins Participating organizations Linnaeus University, Binogi AB, Växjö Municipality Financiers Forte, Växjö Municipality Timetable Oct 2020–Sept 2024 Subject Computer and Information Science (Department of Computer Science and Media Technology, Faculty of Technology)
More about the project
In this doctoral project, we will collaborate with a number of EdTech companies that develop digital learning materials, so we can use their data to drive the development and management of digital methods.
The use of digital learning materials generates various data points (visual/textual) as a part of online learning activities or assignments. However, a common standard for modeling data is lacking. This particular use of data is new also for the different companies producing the digital learning materials, which makes it difficult to replicate the data models.
In this project we aim to organize and structure the collected data to facilitate the interoperability of the digital learning materials, and to make data more understandable and useful for all companies involved, by applying different data mining techniques such as clustering and classification methods. In addition, an entity-relationship diagram and a method of the collected data models will be proposed as an initial standard protocol. The initial standard data model is applicable not only for a single case but for more general datasets, which is really important.
Next, we will propose an analytical workflow to connect methods for data mining, visualization, and machine learning to those of learning progression and trajectories. The next step is to organize different datasets and to analyze the results according to our proposed initial standard protocol. This will be done by applying techniques for visual analysis, machine learning and prediction, in order to understand which visual analysis models are best suited for the data and the users’ needs.
The last step is to develop a visual learning analytic (VLA) dashboard for teachers, to support decision-making and learning trajectories, but also to facilitate and promote pupils' study achievements by combining different individual visualizations for different selected datasets.