Student projects that have achieved international success:

? 2025 | Paper by Kerstin Mayerhofer accepted for CHIIR 2025
The paper "Blending Queries and Conversations" (external link, opens in a new window) by Kerstin Mayerhofer, Rob Capra, and David Elsweiler explores how people interact with a hybrid Web Search + GenAI Chat interface for health information tasks, providing insights into trust, verification, and system choice in search and chat interactions.
The study shows that trust and confidence play a key role in tool selection. In both systems, but particularly when using the chat feature, trust was often misplaced in favour of ease-of-use and seemingly perfect answers, leading to increased confidence post-search despite having incorrect results. Across both tools, 78 distinct search strategies were identified.
The paper is based on Kerstin Mayerhofer's master's thesis.

? 2024 | Paper by Wan-Hua Her accepted for LREC 2024
Together with Udo Kruschwitz,Wan-Hua Her published the paper "Investigating Neural Machine Translation for Low-Resource Languages: Using Bavarian as a Case Study" (external link, opens in a new window), which was accepted for the LREC and published in the "SIGUL2024 Workshop". The paper is based on Wan-Hua Her's master's thesis.

? 2024 | Paper by Maximilian Schmidhuber accepted for LREC 2024
Maximilian Schmidhuber submitted the paper "LLM-Based Synthetic Datasets: Applications and Limitations in Toxicity Detection" (external link, opens in a new window) together with Udo Kruschwitz. The paper was published in "Proceedings of the Fourth Workshop on Threat, Aggression & Cyberbullying @ LREC-COLING-2024". The paper is based on Maximilian Schmidhuber's master's thesis.

? 2024 | Paper by Maximilian Weissenbacher accepted for LREC 2024
Together with Udo Kruschwitz,Maximilian Weissenbacher wrote the paper "Analysing Offensive Language and Hate Speech in Political Discourse: A Case Study of German Politicians" (external link, opens in a new window), which was published in "Proceedings of the Fourth Workshop on Threat, Aggression & Cyberbullying @ LREC-COLING-2024". The paper is based on Maximilian Weissenbacher's master's thesis.

? 2024 | Katharina Buckmayer is now a doctoral candidate at the Técnico University Lisbon, Portugal
Katharina Buckmayer has successfully completed her Bachelor of Arts and Master of Arts in Information Science at the University of Regensburg. She is now a PhD candidate in Digital Media at Técnico University Lisbon and is part of the Inclusive Computing Lab of the Interactive Technologies Institute/LARSyS. Her research interests lie in the design of inclusive, multisensory smart toys and learning environments for young children. She is also concerned with the ethical implications of technology-based research with this particularly vulnerable population group. Her work aims to promote inclusive education and support healthy and positive child development.

? 2023 | Hoai Nam Tran publishes article at the ROMCIR Workshop 2023
The Workshop on Reducing Online Misinformation through Credible Information Retrieval(ROMCIR) was held in conjunction with ECIR 2023. Hoai Nam Tran presented his paper "Towards Reducing Misinformation and Toxic Content Using Cross-Lingual Text Summarisation", which is available as a presentation (external link, opens in a new window) and will be published in the proceedings of the workshop (external link, opens in a new window).

? 2023 | Markus Bink presents paper at CHIIR 2023
Markus Bink presented his paper "Investigating the Influence of Featured Snippets on User Attitudes" (external link, opens in a new window) at the CHIIR 20 (external link, opens in a new window)23, which took place this year in Austin (Texas, USA). The paper was written together with Sebastian Schwarz (Master's student in media informatics) and was supervised by David Elsweiler and Tim Draws (Delft University of Technology).

? 2022 | Paper by Maximilian Weissenbacher presented at KONVENS 2022
Maximilian Weissenbacher'spaper on "Sentiment Analysis on Twitter for the Major German Parties during the 2021 German Federal Election" (external link, opens in a new window) was presented at KONVENS 20 (external link, opens in a new window) 22. The study looked at how the sentiment of political parties changes on Twitter during an election period. It used text classification methods and compared their performances, finding that transformer-based models such as the bidirectional encoder of transformers (BERT) performed better than traditional machine learning models such as Naive Bayes and lexicon-based models such as GerVADER. The paper was written together with Thomas Schmidt (who gave the talk at KONVENS 2022), Jakob Fehle, Jonathan Richter, Philipp Gottschalk and Christian Wolff.

? 2022 | 1st place for Hoai Nam Tran at the CLEF 2022 CheckThat! Lab
Hoai Nam Tran successfully completed Task 3B of the CheckThat! Lab of the CLEF 2022 and took first place with his approach. The topic was "Fighting the COVID-19 Infodemic and Fake News Detection". The focus was on cross-lingual fake news detection and was extremely challenging due to the combination of English training data and German test data, whose extracted news articles comprised up to 100,000 characters. Out of 16 participating teams, only 8 managed a correct submission and even here Hoai Nam Tran was able to clearly (6.5%) outperform the runners-up with a Macro-F1 score of 0.290 and an accuracy of 0.427. His approach consisted of three parts: Summarisation technique, machine translation, and multi-class classification. The results have been published as a paper at CLEF 2022 (external link, opens in a new window).

? 2022 | Paper by Miriam Schirmer accepted for LREC 2022
Together with Gregor Donabauer and Udo Kruschwitz,Miriam Schirmer published the paper"A New Dataset for Topic-Based Paragraph Classification in Genocide-Related Court Transcripts (external link, opens in a new window)", which was accepted for LREC 2022 and also presented as a poster. Miriam Schirmer studied criminology at the UR and attended a course on natural language engineering, which had a lasting impact on her. The paper is also part of her doctorate in computational social science at the Technical University of Munich.

? 2022 | Paper by Philipp Hart l accepted for LREC 2022
Together with Udo Kruschwitz,Philipp Hartl wrote the paper"Applying Automatic Text Summarisation for Fake News Detection (external link, opens in a new window)", which was accepted for LREC 2022. The paper is based on Philipp Hartl's master's thesis.

? 2022 | Paper by Christoph Turban accepted for LREC 2022
Christoph Turban submitted together with Udo Kruschwitz the paper"Tackling Irony Detection using Ensemble Classifiers (external link, opens in a new window)" for LREC 2022. The paper was accepted. This paper is based on Christoph Turban's bachelor's thesis.

? 2022 | Paper by Markus Bink accepted for CHIIR 2022
Markus Bink submitted the paper "Featured Snippets and their Influence on Users' Credibility Judgements" for CHIIR 2022, which was accepted (external link, opens in a new window) for the conference (external link, opens in a new window). The paper was part of his bachelor's thesis and was realised together with Steven Zimmerman and David Elsweiler.

? 2021 | Master's thesis published as a short paper at GIScience 2021
Gregor Donabauer, in collaboration with Prof. Dr Bernd Ludwig, published his master's thesis entitled"Testing Landmark Salience Prediction in Indoor Environments Based on Visual Information (external link, opens in a new window)" as a short paper at GIScience 2021.

? 2021 | Shared task for "GermEval 2021"
Hoai Nam Tran and Maximilian Schmidhuber each successfully participated as an individual team in the Shared Task of GermEval 2021 (external link, opens in a new window). The subtasks posed challenges in recognising and classifying toxic hate comments on the one hand and valuable comments on the other, as well as facts and claims. Hoai Nam Tran took 2nd place overall out of 15 teams. Maximilian Schmidhuber completed the first subtask and took 9th place at . Their performance is particularly noteworthy because they competed as individual student teams against group teams of doctoral students, postdocs and industry representatives. →More information in the proceedings (external link, opens in a new window)

? 2021 | Published paper for SwissText 2021
In collaboration with Prof Dr Kruschwitz, Gregor Donabauer wrote the paper "University of regensburg@ swisstext 2021 sepp-nlg: Adding sentence structure to unpunctuated text" (external link, opens in a new window). It was written as part of a shared task at the SwissText 2021 conference and is related to his presentation and project at WWW'21.

? 2021 | Presentation at the Web Conference 2021
At WWW '21, Gregor Donabauer (together with Udo Kruschwitz and David Coney) gave a talk on "Making Sense of Subtitles: Sentence Boundary Detection and Speaker Change Detection in Unpunctuated Texts (external link, opens in a new window)". This project was part of the seminar "Advanced Topics in Information Retrieval".
? 2021 | Short paper presented by Dominik Ramsauer at the ECIR 2021
Dominik Ramsauer presented a short paper on "Exploring the Incorporation of Opinion Polarity for Abstractive Multi-document Summarisation" (external link, opens in a new window) at ECIR 2021, which took place online. In it, he developed a model for MDS that makes it possible to obtain moods and opinions within abstract summarised documents.
? 2021 | Paper presented by Anna-Marie Ortloff at CHIIR 2021
Anna-Marie Ortloff presented the paper"The Effect of Nudges and Boosts on Browsing Privacy in a Naturalistic Environment (external link, opens in a new window)" at CHIIR 2021, which deals with ways of influencing user behaviour with regard to the disclosure of personal data when surfing the Internet. The publication is the result of her master's thesis - in collaboration with Steven Zimmerman, David Elsweiler and Niels Henze.

? 2021 | Thomas J?nich now a doctoral candidate at the University of Glasgow
Thomas J?nich, a graduate of information science in Regensburg, started his doctorate at the University of Glasgow (external link, opens in a new window) in February 2021. He is working on the topic of "Fair Machine Learning for Search & Recommendation Systems".

? 2021 | Anna-Marie Ortloff now a doctoral student at the University of Bonn
Anna-Marie Ortloff, a former student of information science at the University of Regensburg, started her doctorate at the University of Bonn at the beginning of January 2021. As part of the Behavioural Security (BeSec) (external link, opens in a new window) workgroup under Prof. Dr Matthew Smith, she would like to find her doctoral topic in this area, taking human information behaviour into account.

? 2020/2021 | Hate Speech Detection at EVALITA 2020 and WiDS 2021
Julia Hoffmann (together with Udo Kruschwitz) published the paper "UR NLP @ HaSpeeDe 2 at EVALITA 2020: Towards Robust Hate Speech Detection with Contextual Embeddings" (external link, opens in a new window) at Evalita 2020 (external link, opens in a new window). The paper was written for the conference's shared task, which in turn was initiated in the Master's project course . She also took part in the workshop and the Women in Data Science 2021 (external link, opens in a new window) conference on the topic of "Hate Speech Detection".

? 2020 | Paper by Sabrina Barko-Sherif accepted for CHIIR 2020
Sabrina Barko-Sherif'spaper "Conversational Agents for Recipe Recommendation" (external link, opens in a new window) has been accepted for and presented at CHIIR '20 (The 2020 ACM Conference on Human Information Interaction and Retrieval).

? 2020 | Master's thesis results published
Selina Meyer has published the results of her Master's thesis as an article entitled "Assessing the Quality of Weight Loss Information on the German Language Web" (external link, opens in a new window) in the journal Movement and Nutrition in Health and Disease.

? 2018 | Paper by Anna-Marie Ortloff & fellow students accepted for MuC'18
In September 2018, Anna-Marie Ortloff presented a paper at "Mensch und Computer 2018" in Dresden. This was the result of a seminar paper in the course "Ethics for Software Engineers & Data Scientists". In this paper, it was found that contextual privacy policies, i.e. the presentation of privacy information relevant to the context of use, are accepted by users and that they feel they understand them better than the classic form of presenting privacy policies in continuous text(to the paper (external link, opens in a new window)).
? 2015 | Six student projects accepted for MUM'15
At the end of 2015, six student groups presented their projects for the Master's course "Algorithms for Human-Machine Interaction" at the MUM'15 ("International Conference on Mobile and Ubiquitous Multimedia") in Linz. Fortunately, our students presented six of a total of 15 accepted research projects at this event. Below is a brief overview of the projects:
- "Nombot - Simplify Food Tracking": Nombot is a dialogue system that makes it easier for users to track their eating habits(to the paper).
- "FROY: Exploring Sentiment-Based Movie Recommendations": FROY is a mobile film recommendation system that compares a user's current emotional state with the mood of films(to the paper).
- "Landmark Mining on a Smartwatch Using Speech Recognition": The project dealt with the question of how landmark data for a navigation system can be obtained using a smartwatch and speech recognition(to the paper).
- "Destmaster - Improved Destination Input System for the UR-Walking Application": The project aimed to optimise the input of navigation destinations for the university pedestrian navigation app "URWalking"(to the paper).
- "Emotion-based Music Recommendation using Supervised Learning": The project was dedicated to the development of a music recommendation system that allows the user to select pieces of music according to emotions(to the paper).
- "Augmented Reality-Based Training of the PCB Assembly Process": The project resulted in an augmented reality assistance system that supports industrial workers in the production process of printed circuit boards with the help of smart glass and self-developed software (to the paper).