Open Topics


The Workflow Systems and Technologies Group offers topics for bachelor and master theses as well as Master Praktikum. The following list contains a number of current suggestions for topics.

To discuss one of those or any other topic in this research field, please contact the particulary named supervisor.

» Jump to Bachelor thesis topics

» Jump to Master thesis topics

This list will be updated continuously!

Master thesis topics

 

At the moment there are no open Master thesis topics.

Bachelor theses topics


Supervision Marian Lux

Topic: AI – Using a Large Language Model (LLM) together with Retrieval-Augmented Generation (RAG) for answering questions on custom data

 

Goal:

Develop a basic open source chatbot which can answer questions by using a local LLM, e.g., Llama 2 [1]/Mistral [5] and the RAG approach [2].

The chatbot should be able to answer questions based on data from uploaded documents and websites. The UI should be implemented as web UI (e.g., [3]).

The code will be hosted public on GitHub. The chatbot should handle a particular case and aiming to be production ready by implementing several optimization approaches on. The optimization methods require further literature research. Additionally, finetuning the local LLM – with domain-specific abbreviations by using the QLoRA [5] approach – could be considered if needed.

 

Recommended requirements:

Implementation in Python

GPU with at least 8GB Ram or free Google Colab account

Interest in building LLM AI solutions

 

Supervisor:

Marian LUX - marian.lux@univie.ac.at

Supervision and thesis in German or English

 

References:

[1] https://ai.meta.com/llama/

[2] Lewis, P., Perez, E., Piktus, A., Petroni, F., Karpukhin, V., Goyal, N., ... & Kiela, D. (2020). Retrieval-augmented generation for knowledge-intensive nlp tasks. Advances in Neural Information Processing Systems, 33, 9459-9474.

[3] https://streamlit.io/

[4] https://mistral.ai/news/announcing-mistral-7b/

[5] https://arxiv.org/abs/2305.14314

 

Topic: Explainable AI on Event Logs

 

Goal:

Develop a feature attribution method for a feedforward neural network based on event logs/environmental data from a web app. The data set will be provided. The neural network, containing tabular data, should be developed and trained by using PyTorch [1] or TensorFlow [2]. Feature attribution methods like Integrated Gradients [4] (or a different one [5]) can be applied to make the model explainable. Use literature research to decide between one-hot encoding and embedding on particular features.

A Visualization for feature attribution of selected possible outputs from the trained model by using a modern UI e.g., [3], is expected.

The code will be hosted public on GitHub.

 

Recommended requirements:

Implementation in Python

Using PyTorch or TensorFlow for developing the model

Interest in (explainable) AI

 

Supervisor:

Marian LUX - marian.lux@univie.ac.at

Supervision and thesis in German or English

 

References:

[1] https://pytorch.org/

[2] https://www.tensorflow.org/

[3] https://streamlit.io/

[4] https://arxiv.org/pdf/1703.01365.pdf

[5] https://cgarbin.github.io/machine-learning-interpretability-feature-attribution/#limitations-and-traps-of-feature-attribution

 

Topic: Benchmark for Recurrent Neural Network vs. Deep Feedforward Neural Network

 

Goal:

Develop a recurrent neural network (RNN) and a deep feedforward neural network (FNN) in Python by using PyTorch[1][2] from event logs from a web app together with contextual data like weather forecasts. The data set will be provided. For training the RNN, define an appropriate input sequence length for the provided data set based on literature research. Use literature research to decide between one-hot encoding and embedding on different kinds of data. Also use or develop methods to optimize the hyperparameters for both models.

Finally, the model should be compared with metrics, discovered from literature research. The thesis explains the strengths and weaknesses of both approaches.

The code will be hosted public on GitHub.

 

Recommended requirements:

Implementation in Python

Using PyTorch for developing the model

Interest in AI

 

Supervisor:

Marian LUX - marian.lux@univie.ac.at

Supervision and thesis in German or English

 

References:

[1] https://pytorch.org/

[2] Stevens, E., Antiga, L., & Viehmann, T. (2020). Deep learning with PyTorch. Manning Publications.

[3] Raschka, S., Liu, Y. H., Mirjalili, V., & Dzhulgakov, D. (2022). Machine Learning with PyTorch and Scikit-Learn: Develop machine learning and deep learning models with Python. Packt Publishing Ltd.


Topic: Process Mining Visualization Tool in Python

 

Goal:

Improving/maintaining and providing new features to the already existing process mining tool from a previous bachelor thesis [1]

Different topics based on this process mining tool are possible. Some of them are listed below and the scope of the work is agreed with the supervisor at the beginning. Please write an email with your preferences to the supervisor:

  • Bug fixing for handling a huge amount of input data [1]
  • Changing user interface through Streamlit [2]
  • Enhancing the Heuristic Miner [12] with additional node filtering by using the spm metric [13]
  • Enhancing the Fuzzy Miner with additional filtering and visualization methods [3], [4]
  • Implementing Alpha Miner as additional process mining algorithm [5]
  • Implementing Inductive Miner as additional process mining algorithm [6]
  • Implementing process mining algorithms which act like existing tools [7] [8]
  • Decision tool for which process mining algorithm fits best for a particular use case or data set [3-8] [12-13]
  • Development of an Object Centric Process Mining tool like [9], [10]

The whole work is open source, also the contribution during the bachelor thesis. This means that the code must be easily maintainable and extendable. The written code should contain unit tests as well.

The code will be hosted public on GitHub.

Because of the different topics, multiple students may work simultaneously on the same code base. This implies that at the end of the thesis, all changes must be merged into the main project (GitHub).

 

Recommended requirements:

Implementation in Python

Only basic frameworks (numpy, scikit-learn etc.) can be used. For other more sophisticated frameworks, a permission from supervisor is mandatory.

 

Supervisor:

Marian LUX - marian.lux@univie.ac.at

Supervision and thesis in German or English

 

References:

[1] https://github.com/MLUX-University-of-Vienna/Process-Mining-Visualization

[2] https://streamlit.io/

[3] Okoye, K., Naeem, U., & Islam, S. (2017). Semantic fuzzy mining: Enhancement of process models and event logs analysis from syntactic to conceptual level. International Journal of Hybrid Intelligent Systems, 14(1-2), 67-98.

[4] Günther, C. W., & Van Der Aalst, W. M. (2007, September). Fuzzy mining–adaptive process simplification based on multi-perspective metrics. In International conference on business process management (pp. 328-343). Berlin, Heidelberg: Springer Berlin Heidelberg.

[5] Weerapong, S., Porouhan, P., & Premchaiswadi, W. (2012, November). Process mining using ?-algorithm as a tool (A case study of student registration). In 2012 tenth international conference on ICT and knowledge engineering (pp. 213-220). IEEE.

[6] Leemans, S. J., Fahland, D., & Van Der Aalst, W. M. (2014). Process and deviation exploration with inductive visual miner. In 12th International Conference on Business Process Management, BPM 2014 (pp. 46-50). CEUR-WS. org.

[7] https://www.celonis.com/

[8] https://fluxicon.com/disco/

[9] https://www.ocpm.info/ocel_demo.html

[10] https://encyclopedia.pub/video/video_detail/785

[12] Van Der Aalst, W., & van der Aalst, W. (2016). Data science in action (pp. 3-23). Springer Berlin Heidelberg.

[13] Lux, M., Rinderle-Ma, S., & Preda, A. (2018). Assessing the quality of search process models. In Business Process Management: 16th International Conference, BPM 2018, Sydney, NSW, Australia, September 9–14, 2018, Proceedings 16 (pp. 445-461). Springer International Publishing.

 

Further Literature:

[11] https://research.aimultiple.com/process-mining-algorithms/

 

Topic: Process Mining Framework

 

Goal:

Developing an open-source process mining[1] framework to be used for developing process mining solutions with python, e.g., for visualization tasks in [2].

 

The development of different tools, algorithms (cf., [3,4]) and metrics (e.g., [5]) for this process mining framework are possible. The scope of the work is agreed with the supervisor at the beginning. Please write an email with your preferences to the supervisor:

 

The whole work is open source, also the contribution during the bachelor thesis. This means that the code must be easily maintainable and extendable. The written code should contain unit tests as well.

 

The code will be hosted public on GitHub with an open-source license, like Apache, MIT or BSD.

 

Because of the different topics, multiple students may work simultaneously on the same code base. This implies that at the end of the thesis, all changes must be merged into the main project (GitHub).

 

Recommended requirements:

Implementation in Python

Only basic frameworks (NumPy, pandas etc.) can be used. For other more sophisticated frameworks, a permission from supervisor is mandatory.

Interest in algorithms and/or process mining

 

Supervisor:

Marian LUX - marian.lux@univie.ac.at

Supervision and thesis in German or English

 

References:

[1] Van Der Aalst, W., & van der Aalst, W. (2016). Data science in action (pp. 3-23). Springer Berlin Heidelberg.

[2] https://github.com/MLUX-University-of-Vienna/ProcessMiningVisualization_WS23

[3] https://research.aimultiple.com/process-mining-algorithms/

[4] https://pm4py.fit.fraunhofer.de/

[5] Lux, M., Rinderle-Ma, S., & Preda, A. (2018). Assessing the quality of search process models. In Business Process Management: 16th International Conference, BPM 2018, Sydney, NSW, Australia, September 9–14, 2018, Proceedings 16 (pp. 445-461). Springer International Publishing.


Topic: Individual

 

It is also possible to submit individual topics. Doing so, please submit an abstract to your supervisor via email. If the topic fits the requirements for a bachelor thesis, the scope of the work is agreed with the supervisor at the beginning.

Supervisor:

Marian LUX - marian.lux@univie.ac.at

Supervision and thesis in German or English