Prof. Dr. Stefan Decker - Theses
Master
A Semantic Publish-Subscribe System for the Internet of Production - Open
Supervised by Prof. Dr. Stefan Decker; Advisor(s):
Lars Gleim, M. Sc.
In the context of this master thesis, the student should design and implement the recently introduced SPARQL Subscription specification, enabling the subscription to SPARQL queries and subsequent Pub-Sub-based semantic realtime information systems for the Internet of Production.
Bachelor, Master
An Extensible Web Interface for Linked Data Platform Servers - Open
Supervised by Prof. Dr. Stefan Decker; Advisor(s):
Lars Gleim, M. Sc.,
Dipl.-Medieninf. István Koren
The Linked Data Platform (LDP) is a simple, standardized REST API that allows for the hierarchical organization of Linked Data, similar to a classical file system. While the standard is rather simple, manual interaction with the file system is tiresome and error prone. To simplify the interaction with the LDP server, a suitable web-based user interface for browsing, editing and visualizing LDP data with suitable Web Components should be developed in this work.
Bachelor
Automatic HyperGraphQL Bootstrapping for Triplestores - Open
Supervised by Prof. Dr. Stefan Decker; Advisor(s):
Lars Gleim, M. Sc.,
Dipl.-Medieninf. István Koren
In the context of this bachelor thesis, the student should employ (existing) SPARQL queries to automatically extract the data schema (Ontology) of a given RDF graph and programmatically convert it to a HyperGraphQL Schema, thus greatly simplifying the deployment of GraphQL API endpoints for RDF Triplestores.
Master
Automatic Provenance Generation for Serverless Computing - Open
Supervised by Prof. Dr. Stefan Decker; Advisor(s):
Lars Gleim, M. Sc.
In the context of this work, the student should extend the Function-as-a-Service (FaaS) paradigm of serverless computing with concepts for the automatic tracking of provenance information (information about data origins, uses, actors and involved processes over time, similar to a GIT revision history for data) and implement a corresponding framework using a readily available FaaS platform, such as Kubeless, OpenFaaS or the Fn Project.
Bachelor, Master
Entwicklung einer Co-Simulationsumgebung zur Abbildung des Prozessdatenverkehrs von intelligenten Energienetzen - Open
Supervised by Prof. Dr. Stefan Decker, Dr-Ing. Michael Andres
Bachelor, Master
Entwicklung eines Systems zur Angriffserkennung in intelligenten Energienetzen - Open
Supervised by Prof. Dr. Stefan Decker, Dr-Ing. Michael Andres
Master
Exploring Unknown Environments - Finding Pollution in Underground Pipes - Open
Supervised by Prof. Dr.-Ing. Gerd Ascheid, Prof. Dr. Stefan Decker; Advisor(s):
Dr. Michael Cochez,
Ahmed Hallawa
Bachelor, Master
LD-FUSE: A Filesystem in Userspace for the Linked Data Platform - Open
Supervised by Prof. Dr. Stefan Decker; Advisor(s):
Lars Gleim, M. Sc.
In the context of this work, the student should develop a FUSE user-space file system, allowing to interact with Linked Data Platform servers over HTTP as with a local file system.
Bachelor, Master
Modellierung und Analyse der Anwendbarkeit von Blockchain Technologien in der Energieversorgung - Open
Supervised by Prof. Dr. Stefan Decker, Dr-Ing. Michael Andres
Bachelor
A Distributed Data Revisioning System for the Internet of Production - Running
Supervised by Prof. Dr. Stefan Decker; Advisor(s):
Lars Gleim, M. Sc.
In the context of this thesis, the student should implement a distributed data revisioning system, enabling interorganizational data reuse and extension in the Internet of Production and Internet of Things.
Master
Master
Automated Semantic Type Annotation of Structured Data using Deep learning - Running
Supervised by Prof. Dr. Stefan Decker, Tobias Meisen; Advisor(s):
Lars Gleim, M. Sc.,
André Pomp
Bachelor
Classification of Cancer with methylation aware motifs - Running
Supervised by Prof. Dr. Stefan Decker; Advisor(s):
Dr. Oya Deniz Beyan,
Md. Rezaul Karim, M.Sc.
DNA methylation data has become a popular choice as features for cancer classification tasks. However, a big problem arising here is its high dimensionality in combination with the usually small amount of available samples. Therefore, we created lower-dimensional meta-features that take the effects of the altered DNA methylation pattern on the binding affinity of human Transcription Factors into account. Their performance is evaluated in this thesis.
Bachelor
Datenanalyse zur Messung der Gleichmäßigkeit des Präparationsauftrags im Schmelzspinnprozess - Running
Supervised by Prof. Dr. Stefan Decker, Thomas Gries; Advisor(s):
Dr. Sandra Geisler,
Lukasz Debicki
Bachelor
Dynamic Embeddings of Evolving Knowledge Graphs - Running
Supervised by Prof. Dr. Stefan Decker; Advisor(s):
Dr. Michael Cochez,
Dr. Florian Lemmerich
The goal of this Bachelor thesis is the research of updating KG embeddings with new information in order to obtain a dynamic and stable embedding of the fast-evolving KG while reducing the computational effort.
Fraunhofer ICT Dissertation Award for Dr. Erion Elmasllari 