Knowledge Enriched Natural Language Processing
In the field of Knowledge-Enriched NLP, we work on current topics of Natural Language Processing. Specifically, we are adapting and improving large language models (LLMs) such as BERT and its derivatives. Our particular focus lies in incorporating explicit knowledge, such as knowledge graphs.
Our application areas range from analyzing historical literature (where current language models struggle due to the length of the texts) to product reviews, and even to unconventional media forms for NLP, such as comments on http://twitch.tv. These media forms present their own challenges due to their unique language style. In addition to analyzing pure text, we also investigate the adaptability of NLP methods for processing mathematical equations.
In projects like Kallimachos or CLiGS we collaborate with literary scholars and work on literary and NLP research questions. In MOTIV, we work with psychologists to analyse the interaction between users and smart devices.
Core Research Topics include:
- Sentiment Analysis
- Detection of scenes (i.e., coherent segments in literary texts)
- Combination of Knowledge Graphs with NLP methods
- Adaption of large language models
- NLP-based models for recommendation
People
Projects
-
Detecting Scenes in Fiction - Breaking down long literary texts from novels into coherent segments.
-
MOTIV - Analyising interactions with language-based interactive technology and AI regarding digital interaction literacy.
-
DZ.PTM - Building an knowledge-based service management in radiology with the University Clinic Würzburg.
-
Knowledge Graphs - Combining neural networks and knowledge graphs.
-
Twitch - Analysing messages from twitch.tv regarding sentiment and other factors.
-
KILiMod - Machine learning based chat moderation and content enrichment
Concluded Projects
-
Kallimachos - Building a complete text analysis pipeline, starting with OCR from paper and going up to high-level text mining.
-
CLiGS - CLiGS combines large text collections with innovative analysis methods and hermeneutic sensibility for context.
-
ModernGBERT: German-only 1B Encoder Model Trained from Scratch. . 2025.
-
{O}tterly{O}bsessed{W}ith{S}emantics at {S}em{E}val-2024 Task 4: Developing a Hierarchical Multi-Label Classification Head for Large Language Models. . In Proceedings of the 18th International Workshop on Semantic Evaluation (SemEval-2024), A. K. Ojha, A. S. Do{\u{g}}ru{\"o}z, H. Tayyar Madabushi, G. Da San Martino, S. Rosenthal, A. Ros{\’a} (eds.), pp. 602–612. Association for Computational Linguistics, Mexico City, Mexico, 2024.
-
LL\"aMmlein: Compact and Competitive German-Only Language Models from Scratch. . 2024.
-
BibSonomy Meets ChatLLMs for Publication Management: From Chat to Publication Management: Organizing your related work using BibSonomy & LLMs. . 2024.
-
The {F}airy{N}et Corpus - Character Networks for {G}erman Fairy Tales. . In Proceedings of the 5th Joint SIGHUM Workshop on Computational Linguistics for Cultural Heritage, Social Sciences, Humanities and Literature, pp. 49–56. Association for Computational Linguistics, Punta Cana, Dominican Republic (online), 2021.
-
Detecting Scenes in Fiction: A new Segmentation Task. . In Proceedings of the 16th Conference of the {E}uropean Chapter of the Association for Computational Linguistics: Volume 1, Long Papers. ACL, 2021.
-
Shared Task on Scene Segmentation @ KONVENS 2021. . In Shared Task on Scene Segmentation @ KONVENS 2021, pp. 1–21. 2021.
-
Emote-Controlled: Obtaining Implicit Viewer Feedback through Emote based Sentiment Analysis on Comments of Popular Twitch.tv Channels. . In ACM Transactions on Social Computing. 2020.
-
HarryMotions – Classifying Relationships in Harry Potter based on Emotion Analysis. . In 5th SwissText & 16th KONVENS Joint Conference. 2020.
-
LM4KG: Improving Common Sense Knowledge Graphs with Language Models. . In International Semantic Web Conference. Springer, 2020.
-
Detection of Scenes in Fiction. . In Proceedings of Digital Humanities 2019. 2019.
-
Analysing Direct Speech in German Novels. . In DHd 2018. 2018.
-
Burrows’ Zeta: Exploring and Evaluating Variants and Parameters. . In DH, pp. 274–277. 2018.
-
A White-Box Model for Detecting Author Nationality by Linguistic Differences in Spanish Novels. . In DH. ADHO, 2018.
-
Burrows Zeta: Varianten und Evaluation. . In DHd 2018. 2018.
-
Towards Sentiment Analysis on German Literature. . 2017.
-
Prediction of Happy Endings in German Novels. . In Proceedings of the Workshop on Interactions between Data Mining and Natural Language Processing 2016, P. Cellier, T. Charnois, A. Hotho, S. Matwin, M.-F. Moens, Y. Toussaint (eds.), pp. 9–16. 2016.
-
Classification of Literary Subgenres. . In DHd 2016. 2016.
-
Straight Talk! Automatic Recognition of Direct Speech in Nineteenth-Century French Novels. . In DH, pp. 346–353. 2016.
-
Analyzing Features for the Detection of Happy Endings in German Novels. . 2016.
-
Straight Talk! Automatic Recognition of Direct Speech in Nineteenth-Century French Novels. . In DH, pp. 346–353. 2016.
-
Significance Testing for the Classification of Literary Subgenres. . In DH 2016. 2016.
-
Genre classification on German novels. . In Proceedings of the 12th International Workshop on Text-based Information Retrieval. 2015.
-
Proceedings of the 1st International Workshop on Interactions between Data Mining and Natural Language Processing co-located with The European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, DMNLP@PKDD/ECML 2014, Nancy, France, September 15, 2014. . In Vol. 1202 of {CEUR} Workshop Proceedings. CEUR-WS.org, 2014.