Home // Education // Theses

Our Research Topics

Are you still looking for a topic or a possible supervisor for your thesis? See what the researchers of ScaDS.AI Dresden/Leipzig are currently working on!

Theses

We welcome students, for example from the T UD Dresden University of Technology or the Leipzig University, to write their theses at ScaDS.AI Dresden/Leipzig. You should have an interest in:

Data Science,
Artificial Intelligence,
Big Data, or
other topics related to the research field of Computer Science.

In addition to master’s and bachelor’s theses, semester theses might also be possible depending on your prefered topic. Most final theses that are offered can be written and supervised in German as well as in English. If you don’t find a suitable topic in this list, feel free to send us a message with your research interests and initial ideas for your thesis. In case you need an inspiration for your topic, you might want to have a look at the different research topics at our center. We will support you in finding a supervisor if you have already decided on a suitable topic. Our team is happy to answer your questions about writing a final thesis at ScaDS.AI Dresden/Leipzig.

About us

Research

Education

Transfer and Service

Living Lab

Our Research Topics

Theses

Detecting and Mitigating Hate Speech with Large Language Models: Evaluating Strategies and Ethical Considerations

Evaluating Knowledge Retrieval in Multi-Agent Systems with Large Language Models

Automated evaluation of scintigraphic images of horses

Exploring Citation Generation with Large Language Models: A Comparative Study of Prompting Strategies

GPT-3, BERT & Co.: When to use which language model?

Using Quantum Computing in Natural Language Processing

Alles nur Show? Ein automatischer Vergleich von Nachrichten vor der Bundestagswahl mit dem Koalitionsvertrag mittels Natural Language Processing

Creating a Dataset of Textual Descriptions of Planned Actions and Expectations of Entities

Creating a Multimodal Dataset of Actions Labeled with their Expected Durations

Creating a Dataset of Complex Temporal Questions for Testing Large Language Models (LLMs)

Meta hyperparameter tuning for multi-omics data integration by autoencoders

Exploring Methods and Technologies in our Living Lab Environment

Advanced Multi-Modality Learning in Electronic Health Records for Personalized Medical Recommendations

Design and Development of a Speech Collection Software for Spoken Natural Languages

Fusing RDF Knowledge Graphs with Deep Learning for Advanced Recommender Systems

Extending the RDF Knowledge Graph SemOpenAlex.org

Large Language Model-enhanced Graph Message Passing Network for Link Prediction

Stock Market Predictions through Deep Learning

Designing and Executing a Large-Scale User Study on Scientific Text Simplification

Morphologically Biased Byte-Pair Encoding

Comparative analysis of CNN models for semantic segmentation of segetal flora using UAV imagery from different flight heights

Manual and Automated Detection of Bias in Medical Data of ICUs

Automated Question Generation from Wikipedia and Wikidata

Latency and Cost Analysis of Prompt Compression for LLM’s

Biomechanical User Models for Mobile Device Control based on Reinforcement Learning

Leveraging Large Language Models to Analyze the Public Discourse on Driving Automation Shortcomings in Social Platforms

Content-Adaptive Depth Aware Image Restoration

Temporal Ordering of Free-from Text Using Weighted Date Extraction

Detection of Repetitive Patterns in Time Series - Applied on Systems Monitoring Data

Opening the Black Box of Machine Learning

Verbesserung der Erkennung interessanter Punkte in Standortverläufen durch Analyse der Semantik

Privatsphäreanalyse von NextBike-Fahrradleihdaten

Anomaly detection in geodata with AutoEncoders

Entwicklung einer Android App zur Messung und Visualisierung des Privatsphäre-Leaks durch Standortabfragen von Smartphone-Apps

Applying a reinforcement learning framework in a testing environment

Methods to Cross the simulation-to-reality gap on a Autonomous Driving Model Car

Data augmentation of time series data

Netzwerkerkennung und -reaktion auf Bedrohungen für Smart-Home-Geräte mittels statistischer Klassifikationsverfahren

Investigating Training an Agent to Drive a Vehicle in a Simulated Environment Using Reinforcement Learning

Pose Classification in an Omnidirectional Camera Setup using Transfer Learning and Open Source Models and Datasets

Resource and energy consumption of transformers-based omics language models

Denoising Cryoseismological Distributed Acoustic Sensing Data Using a Convolutional Neural Network

Comparing Anomaly-Based Network Intrusion Detection Approaches Under Practical Aspects

Entwicklung eines HIDS auf Basis des LID-DS unter Verwendung einer Self-Organizing Map

Klassifikation von Bilddaten mithilfe von PDG-inspirierten neuronalen Netzen in der additiven Fertigung

Privacy-Preserving Variational Autoencoder for Single-Cell Analysis

Abstract Argumentation in Artificial Intelligence - Modularization and its Consequences for Labeling-based Semantics

Defining Defense in Abstract Argumentation from Scratch – A Generalizing Approach

An Implementation of Recursive Semantics in Abstract Argumentation Frameworks

Über die durchschnittliche Anzahl vollständiger Extensionen in abstrakten Argumentationssystemen – Theorie und Implementierung

Choices and their Consequences - On Decompositions of Complete Extensions in Argumentation Frameworks

Benchmarking of federated learning tools for genetic data analysis

Model training of a simulated self-driving vehicle using an evolution-based neural network approach

Demonstrator für Zufallszahlen in Hard- und Software

Temporal Graph Metrics

Implementierung und Evaluation von parallelen Verfahren mit Flink zum Schutz personenbezogenener Daten beim Record Linkage

Storage patterns for data-local processing of georeferenced data in BigData systems

Entwicklung von Techniken zur Datenintegration und Datenqualitätsverbesserung für die Graph-Processing-Platform GraDoop

Verwaltung und Verarbeitung von Massenspektrometerdaten

Vergleich und Evaluation von RDF-on-Hadoop Lösungen

Performance Measurement Preparation with AspectJ

Parallelisation of a Hierarchical Volume Data Preprocessor

Eignung von Workflow-Management-Tools für BigData-Aufgabenstellungen

Parallel Metric Space für Privacy Preserving Record Linkage

Efficient storage and retrieval of temporal graph data

Integration of Graph-Analytics Operators within the KNIME-Analytics Platform

Duplicatedetection and Linking in Big Data Analytics Workflows

Linking and Duplicate Detection in Big Graphs - Similarity Measures

Server-Side Aggregation of Time-Series Data in Distributed NoSQL Databases

Vergleich und Evaluation von SQL-on-Hadoop Lösungen

Evaluation and Comparison of Big Data ETL Frameworks

Apache Beam and the price of unification

Erzeugung und Statistische Bewertung von Multigenomgraphen