Writing a Masterthesis

  • Attend our lectures to find out about relevant topics for a master thesis
  • Ask the people in the group about their research and possible topics
  • Search and find concrete topics on our webpage (regularly updated)
  • Bring your own ideas, as long as they fit with our expertise, they are more than welcome

  • At the time of registration, topic, supervisor, and starting date have to be fixed with our secretary
  • Please appoint both your topic and your starting date with your supervisor prior to registration
  • Make sure that you have no other exams or obligations within the time of writing the thesis
  • Make sure that you attended the course "Scientific Writing" before starting

  • Your thesis can be written in English or German, we recommend English
  • Your thesis has to be written in Latex, a document template is provided
  • Regular meetings with the supervisor (at least once a month) are mandatory
  • A fundamental part has to be submitted after 2 months

  • The thesis must be finished within six months
  • It should have a length between 80 and no more than 120 pages
  • Hand in three copies, software, data, and source code have to be submitted on CD-ROM, together with instructions for installing and running the software
  • Your thesis has to be submitted to our secretary, stick to the opening hours
  • Extensions can only be granted in exceptional cases

Topics for Master Theses

The AI chair of Prof. Stuckenschmidt offers topics from the areas listed in the following. For the other chairs of the DWS group (a selection of) concrete topics are presented below.

Smart Process Technologies

Today the majority of data available is produced by processes in terms of log files from software systems and readings from sensors that are observing processes in the physical world. We want to develop technologies that use this data as well as explicit process models to improve process performance and support users involved in the process. On the technical side, topics of interest include the generation of process models from log data (process mining), the validation and comparison of process models as well as the prediction of process behavior. We are further interested in case studies on supporting business and also personal processes using process data and models. Candidates will either adapt and apply data analysis methods to real world processes or develop case studies in process optimization and guidance. We expect basic programming skills and knowledge in business process management and corresponding modelling languages.

Interested Students should contact Professor Stuckenschmidt (heiner(at)informatik.uni-mannheim.de) or Dr. Christian Bartelt (bartelt(at)es.uni-mannheim.de) for further Information.

Decision Support Technologies

Decision Making under Uncertainty is a common requirement in a Business environment. Decision Theory has been developed as a formal framework for supporting such decision by quantifying uncertainty and risks and providing methods for computing expected benefits of decisions. We are interested in implementing and applying methods from decision theory to problems in data analysis and information management. Candidates will typically investigate the state of the art in logical and probabilistic methods for decision theory, implement or adapt an existing method and apply it to a given decision problem. We expect solid programming skills in JAVA, knowledge in probability theory and statistics as well as basic knowledge of formal logic.

Interested Students should contact Professor Stuckenschmidt (heiner(at)informatik.uni-mannheim.de) or Dr. Melisachew Chekol (mel(at)informatik.uni-mannheim.de) for further Information.

Information Management and Mining for the Finance Sector

In the financial sector, investment decisions are typically based on detailed information about sectors, markets and companies. Relevant information includes official reports and key performance indicators as well as analysts’ reports and business news. In order to support investment decisions, this information has to be found, integrated, managed and analyzed. We offer theses on topics in financial information management and mining, whose main focus will be on conducting studies on: a) extracting and interpreting financial data from heterogeneous sources, in particular news feeds and business reports; b) using this information to predict financial trends and behaviors. Candidates will typically investigate the state of the art in information extraction and mining, implement and adapt suitable methods and apply them to a relevant information source or case study. We expect solid programming skills in Java and knowledge in information retrieval or text analytics as well as a basic understanding of Semantic Web Technologies.

Interested Students should contact Professor Stuckenschmidt (heiner(at)informatik.uni-mannheim.de) or Professor Ponzetto (simone(at)informatik.uni-mannheim.de) for further Information.

IT Service Management

IT service management (ITSM) is concerned with the implementation of quality IT services that meet the needs of customers, and is performed by the IT service provider through an appropriate mix of people, process and information technology. In cooperation with the RUM – the main provider of IT services for the University of Mannheim – we offer different topics for master theses that are concerned with the improvement of service provision. Candidates will typically investigate existing structures and processes at the RUM, compare them to the state of the art and other organizations and propose improvements. We expect some knowledge in (IT-) Management and Organization as well as basic knowledge about modern IT infrastructures.

Interested Students should contact Professor Stuckenschmidt (heiner(at)informatik.uni-mannheim.de) for further Information.

Available, ongoing and completed Masterthesis

A large number of e-shops have started to markup structured data about products and offers in their HTML pages using the markup standard Microdata and...

more

Since mid-September 2015, the threat from ransomware has grown considerably [1]. Against this background, comprehensive geographical and temporal...

more

Target: Master

Type: Survey

Short abstract: This thesis should provide an in-depth overview of the adoption of natural language processing and...

more

Target: Master

Type: Experiments

Introduction/problem:  Party manifestos (https://manifestoproject.wzb.eu/) present the vision of a specific party...

more

Title: Enhancing Domain Specific Entity Linking

Target: Master

Type: Experiments

Introduction/problem: Entity linking could improve text...

more

In recent years thanks to the success of social networks (e.g., Facebook, Twitter, ...) and the availability of collaborative customer-made company...

more

A large number of e-shops have started to markup structured data about products, offers and reviews in their HTML pages using the markup standard Micr...

more

Data integration problems arise whenever data from separate sources needs to be combined as the basis for new applications. Within the context of the...

more

This thesis should provide an in-depth overview of the various recurrent neural network models (fully recurrent networks, recursive networks, long...

more

This thesis should provide an in-depth overview of the state-of-the-art methods for representing knowledge graphs and knowledge bases in the (i.e.,...

more