Quick Overview

The goal of GraphIT is to develop and evaluate a dependency graph of learning contents. This graph enables faculties to develop better study programs, teachers to structure their courses better and students to find the best way to achieve their personal goals.

💡 On each interactive visualization on this page, you can click on the "GraphIT Query Service" link in the lower left corner. This opens a new tab which shows you the SPARQL query and allows you to modify it.

The Dependency Graph

Learning contents connect to each other to form a learning path

A dependency graph is a collection of nodes (or items), connected with each other using edges (or properties), to show which nodes depend on others. For GraphIT the nodes represent a topic of study.

At the lowest level our graph is structured by linking learning contents to their prerequisite(s). Students can use these links to determine how topics depend on each other, and what knowledge is required to understand a specific item. Materials (such as tutorials and articles) or exams can also be linked to each learning content.

In the end the graph turns into a big interconnected structure:

A snippet of the dependency graph

In order to make this amount of data useable, several Structure-Items have been defined to help group items together. For example learning contents of similar topics are included in a category, that can later be referenced when putting a course together. Courses include Session-Items (representing an actual session of a course), that also include all items that are being discussed in person.

Items being discussed in a session on 3D-Graphics of the ''CGBV''-course

The User-Item of a student and all items connected to it

Every item can further include existing properties. This allows for specifying information such as a date, expressing interest, people responsible and URLs.

People can interact with the graph by creating a User-Item (example: Max Mustermann), that they can use to link back to items in different ways.

For a more detailed explanation of the entire underlying structure go to GraphStructure or take a quick look at these following diagrams:

• 

  The base structure

• 

  Connection of courses

Using the Graph

The graph itself is just a versatile data structure. In order to make it usable, we are developing tools and visualizations. The SPARQL query language offered by WikiBase allows for retrieving relevant parts of the graph and visualizing it in different ways. This can offer insights on about the contents of the graph, but also the graph itself and it's underlying structure.

We combine different visualization to create a course page (see for example: WissArb and CGBV), where a collection of information is contained in one place.

Such as a timeline of all course sessions:

All sessions of the CGBV course

Or an overview of students knowledge and interests:

Students self defined interests and prior knowledge for learning contents of the WissArb-Course

Such a graph is possible by allowing students to modify their personal user-item and use it to link to items with the properties: has completed and interested in.

Or learning paths, to show the step-by-step path towards learning a specific topic:

The direct learning path to Q149

Custom Frontends

A visual editor for graph exploration and marking items

A tabular editor for modifiying and connecting items

For an easier and more direct way of interaction some custom frontends have been developed.

Students can interact with the graph using a visual editor. It shows a specific course and all items it includes. It allows students to explore the graph, see learning paths towards specific topics and quickly mark items as prior knowledge or a topic that they are interested in.

Admins can interact with a tabular editor, that can modifiy items dependently and independently of each other. They can move, copy or delete statements (property-item-pairs) between items and create new connections.

Plans for the Future

We want to implement:

• Evaluation and implementation of a suitable infrastructure (based on Wikibase)
• Structuring the content of a course of study as a dependency graph
• Continuous evaluation of the usage through pilot projects in courses, in the planning/revision of a degree program and for individual learning paths, e.g. for career changers
• Integration of OER
• Analysis of the course of study aggregated and in person
• Development of tools for visualizing and processing the data
• Preparation for scaling the concept to university / national / international level

Cooperation with other partners is planned, in particular with developers in the area of Wikibase / graph databases and application partners from the public sector.