Categories
Pages
-

DBIS

Text Mining Techniques for Student Assessments

December 14th, 2021

The goal of this thesis is to integrate the open-source framework ReaderBench into the Kubernetes cluster of the tech4comp project. The thesis should investigate related work in the domain of text mining, natural language processing and educational data mining. The developed extension should integrate into the existing project’s infrastructure and provide mentees and their mentors/tutors with advanced feedback on the mentee’s reading comprehension.

Thesis Type Master
Student Karl Yann Zeufack
Status Finished
Proposal on 20/03/2021 12:00 am
Proposal room Online
Presentation on 09/12/2021 12:00 am
Presentation room Online
Supervisor(s) Ralf Klamma
Advisor(s) Alexander Neumann

In our efforts to support heterogeneous communities with the tools and structures they need, we developed our flagship peer-to-peer community platform las2peer. We currently use las2peer in a large research project to extract Learning Analytics data from various Learning Management Systems (e.g., Moodle) with a las2peer connector node and send it via end-to-end encrypted messages through the peer-to-peer network to a network of Learning Record Stores. From there, the data is forwarded to various analytics and mentoring tools, both developed by us and project partners, that analyze the data further to create the information needed for both mentor and mentee support. We currently concentrate on text analysis, using model-based tools for knowledge assessment (e.g. T-MITOCAR).

To take this process further and provide even more insights for the above mentioned target groups, one promising service to integrate is the open-source framework ReaderBench, which allows text mining techniques, advanced natural language processing, and social network analysis for text comprehension exercises.

The goal of this thesis is to integrate the open-source framework ReaderBench into the Kubernetes cluster of the tech4comp project. The thesis should investigate related work in the domain of text mining, natural language processing and educational data mining. The developed extension should integrate into the existing project’s infrastructure and provide mentees and their mentors/tutors with advanced feedback on the mentee’s reading comprehension.

If you are interested in this thesis, please do not hesitate to send a message to lange@dbis.rwth-aachen.de.