Tobias Koopmann, M.Sc.

Chair of Data Science (Informatik X)
University of Würzburg
Am Hubland
97074 Würzburg
Germany

Email: koopmann@informatik.uni-wuerzburg.de

Phone: (+49 931) 31 - 89363

Office: Room B100 (Computer Science Building M2)

Fingerprint: CBC7 EACE BBF3 8CF8 EB3F 3478 A5FB FAA7 8F34 AD4E

Projects and Research Interests

I joined the DMIR group for my PhD studies after receiving my masters degree in Computer Science at the university of Würzburg in early 2019. At first, i was working on click trail analysis and human behavior prediction in the web based on Wikipedia click trails.

I started to work in the REGIO project. My contribution in the project is to analyse locally successful research cooperations based on different properties of the co-author network. For this we leverage the Bayesian approach HypTrails and apply it on graph- structured data. End 2021 after the project has been successfully finalized, I started working on the HydrAS project, which is a natural succession for my research topic.

Current Thesis/Practica Topics

Recommendation of Co-Authorship using Hypergraph Networks

Recently a new field of Graph Neural Networks has been revealed, which takles the field of Hypergraph networks. These networks use hyperedges, which can connect mutiple vertexes. A bibliometric dataset contains authors as vertices and co-occurences on publications as hyperedges. Given this dataset and the method of Hypergraph networks, we can now train a network to predict co-authorship.

Feng, Y., You, H., Zhang, Z., Ji, R., & Gao, Y. (2019, July). Hypergraph neural networks. In Proceedings of the AAAI Conference on Artificial Intelligence (Vol. 33, No. 01, pp. 3558-3565).

Options: Ma Praktika, Ma Thesis

Analysis of Transformer Networks with respect to Layer Dropout and Dataset Splits

Recent work hast proposed the transformer architecture. Other work analysed this architecture regarding its sensitivity on dataset splits with respect to the dataset length.
The main task of this work is to evaluate the transformer and/or BERT architecture with respect to dataset splits not only dependant of the length.

Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., ... & Polosukhin, I. (2017). Attention is all you need. Advances in neural information processing systems, 30.

Csordás, R., Irie, K., & Schmidhuber, J. (2021). The devil is in the detail: Simple tricks improve systematic generalization of transformers. arXiv preprint arXiv:2108.12284.

Options: Ba Thesis, Ma Thesis

Implementation of Automatic HypTrails Framework

The goal of this work is to implement a Python framework, which automatically creates all possible hypothesis and calculates their respective evidences according to HypTrail[1]. The input for this framework is supposed to be an attributed multigraph[2] and some distance measures for each attribute of the nodes.

Possible thesis: BA, MP, MA

[1] Singer, P., Helic, D., Hotho, A. & Strohmaier, M. (2015). HypTrails: A Bayesian Approach for Comparing Hypotheses About Human Trails on the Web. Proceedings of the 24th International Conference on World Wide Web (p./pp. 1003--1013), New York, NY, USA: ACM. ISBN: 978-1-4503-3469-3

[2] Espín-Noboa, L., Lemmerich, F., Strohmaier, M. et al. JANUS: A hypothesis-driven Bayesian approach for understanding edge formation in attributed multigraphs. Appl Netw Sci 2, 16 (2017). https://doi.org/10.1007/s41109-017-0036-1

Teaching

Winter term 22/23:

Datenbanken

ummer term 22:

Seminar: Ausgewählte Themen des Machine Learning

Winter term 21/22:

Seminar: Ausgewählte Themen des Machine Learning

Summer term 21:

Seminar: Ausgewählte Themen des Machine Learning

Winter term 20/21:

Datenbanken

Summer term 20:

Summer term 19:

Praktikum: Natural Language Processing

Winter term 18/19:

Datenbanken

Publications

2021[ to top ]

Koopmann, T., Kobs, K., Herud, K., Hotho, A. (2021) CoBERT: Scientific Collaboration Prediction via Sequential Recommendation, in 2021 International Conference on Data Mining Workshops (ICDMW), 45–54, available: https://doi.org/10.1109/ICDMW53433.2021.00013.
- [ Abstract ]
- [ BibTeX ]
- [ Download ]
Collaborations are an Important factor for scientific success, as the joint work leads to results individual scientists cannot easily reach. Recommending collaborations automatically can alleviate the time consuming and tedious search for potential collaborators. Usually, such recommendation systems rely on graph structures modeling co-authorship of papers and content-based relations such as similar paper keywords. Models are then trained to estimate the probability of links between certain authors in these graphs.In this paper, we argue that the order of papers is crucial for reliably predicting future collaborations, which is not considered by graph-based recommendation systems. We thus propose to reformulate the task of collaboration recommendation as a sequential recommendation task. Here, we aim to predict the next co-author in a chronologically sorted sequence of an author’s collaborators. We introduce CoBERT, a BERT4Rec inspired model, that predicts the sequence’s next co-author and thus a potential collaborator. Since the order of co-authors of a single paper is not that important compared to the overall paper order, we leverage positional embeddings encoding paper positions instead of co-author positions in the sequence. Additionally, we inject content features about every paper and their co-authors. We evaluate CoBERT on two datasets consisting of papers from the field of Artificial Intelligence and the journal PlosOne. We show that CoBERT can outperform graph-based methods and BERT4Rec when predicting the co-authors of the next paper. We make our code and data available.

@inproceedings{koopmann2021cobert, abstract = {Collaborations are an Important factor for scientific success, as the joint work leads to results individual scientists cannot easily reach. Recommending collaborations automatically can alleviate the time consuming and tedious search for potential collaborators. Usually, such recommendation systems rely on graph structures modeling co-authorship of papers and content-based relations such as similar paper keywords. Models are then trained to estimate the probability of links between certain authors in these graphs.In this paper, we argue that the order of papers is crucial for reliably predicting future collaborations, which is not considered by graph-based recommendation systems. We thus propose to reformulate the task of collaboration recommendation as a sequential recommendation task. Here, we aim to predict the next co-author in a chronologically sorted sequence of an author’s collaborators. We introduce CoBERT, a BERT4Rec inspired model, that predicts the sequence’s next co-author and thus a potential collaborator. Since the order of co-authors of a single paper is not that important compared to the overall paper order, we leverage positional embeddings encoding paper positions instead of co-author positions in the sequence. Additionally, we inject content features about every paper and their co-authors. We evaluate CoBERT on two datasets consisting of papers from the field of Artificial Intelligence and the journal PlosOne. We show that CoBERT can outperform graph-based methods and BERT4Rec when predicting the co-authors of the next paper. We make our code and data available.}, author = {Koopmann, Tobias and Kobs, Konstantin and Herud, Konstantin and Hotho, Andreas}, booktitle = {2021 International Conference on Data Mining Workshops (ICDMW)}, keywords = {recommendation}, month = {dec}, pages = {45-54}, title = {CoBERT: Scientific Collaboration Prediction via Sequential Recommendation}, year = 2021 }
Koopmann, T., Stubbemann, M., Kapa, M., Paris, M., Buenstorf, G., Hanika, T., Hotho, A., Jäschke, R., Stumme, G. (2021) Proximity dimensions and the emergence of collaboration: a HypTrails study on German AI research, Scientometrics, available: https://doi.org/10.1007/s11192-021-03922-1.
- [ Abstract ]
- [ BibTeX ]
- [ URL ]
- [ Download ]
Creation and exchange of knowledge depends on collaboration. Recent work has suggested that the emergence of collaboration frequently relies on geographic proximity. However, being co-located tends to be associated with other dimensions of proximity, such as social ties or a shared organizational environment. To account for such factors, multiple dimensions of proximity have been proposed, including cognitive, institutional, organizational, social and geographical proximity. Since they strongly interrelate, disentangling these dimensions and their respective impact on collaboration is challenging. To address this issue, we propose various methods for measuring different dimensions of proximity. We then present an approach to compare and rank them with respect to the extent to which they indicate co-publications and co-inventions. We adapt the HypTrails approach, which was originally developed to explain human navigation, to co-author and co-inventor graphs. We evaluate this approach on a subset of the German research community, specifically academic authors and inventors active in research on artificial intelligence (AI). We find that social proximity and cognitive proximity are more important for the emergence of collaboration than geographic proximity.

@article{koopmann2021proximity, abstract = {Creation and exchange of knowledge depends on collaboration. Recent work has suggested that the emergence of collaboration frequently relies on geographic proximity. However, being co-located tends to be associated with other dimensions of proximity, such as social ties or a shared organizational environment. To account for such factors, multiple dimensions of proximity have been proposed, including cognitive, institutional, organizational, social and geographical proximity. Since they strongly interrelate, disentangling these dimensions and their respective impact on collaboration is challenging. To address this issue, we propose various methods for measuring different dimensions of proximity. We then present an approach to compare and rank them with respect to the extent to which they indicate co-publications and co-inventions. We adapt the HypTrails approach, which was originally developed to explain human navigation, to co-author and co-inventor graphs. We evaluate this approach on a subset of the German research community, specifically academic authors and inventors active in research on artificial intelligence (AI). We find that social proximity and cognitive proximity are more important for the emergence of collaboration than geographic proximity.}, author = {Koopmann, Tobias and Stubbemann, Maximilian and Kapa, Matthias and Paris, Michael and Buenstorf, Guido and Hanika, Tom and Hotho, Andreas and Jäschke, Robert and Stumme, Gerd}, journal = {Scientometrics}, keywords = {regio}, title = {Proximity dimensions and the emergence of collaboration: a HypTrails study on German AI research}, year = 2021 }

2020[ to top ]

Stubbemann, M., Koopmann, T. (2020) The German and International AI Network Data Set, available: https://doi.org/10.5281/zenodo.3693604.
- [ BibTeX ]
@article{koopmann-germanai_2020, author = {Stubbemann, Maximilian and Koopmann, Tobias}, keywords = {regio}, month = {mar}, publisher = {Zenodo}, title = {The German and International AI Network Data Set}, year = 2020 }
Kobs, K., Koopmann, T., Zehe, A., Fernes, D., Krop, P., Hotho, A. (2020) Where to Submit Helping Researchers to Choose the Right Venue, in Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing: Findings, Online: Association for Computational Linguistics, 878–883, available: https://www.aclweb.org/anthology/2020.findings-emnlp.78.
- [ BibTeX ]
- [ URL ]
@inproceedings{kobs-etal-2020-submit, address = {Online}, author = {Kobs, Konstantin and Koopmann, Tobias and Zehe, Albin and Fernes, David and Krop, Philipp and Hotho, Andreas}, booktitle = {Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing: Findings}, keywords = {regio}, month = {nov}, pages = {878–883}, publisher = {Association for Computational Linguistics}, title = {Where to Submit Helping Researchers to Choose the Right Venue}, year = 2020 }

2019[ to top ]

Koopmann, T., Dallmann, A., Hettinger, L., Niebler, T., Hotho, A. (2019) On the Right Track! Analysing and Predicting Navigation Success in Wikipedia, in Proceedings of the 30th ACM Conference on Hypertext and Social Media, HT ’19, Hof, Germany: ACM, 143–152, available: https://doi.org/10.1145/3342220.3343650.
- [ BibTeX ]
- [ URL ]
- [ Download ]
@inproceedings{Koopmann:2019:RTA:3342220.3343650, address = {New York, NY, USA}, author = {Koopmann, Tobias and Dallmann, Alexander and Hettinger, Lena and Niebler, Thomas and Hotho, Andreas}, booktitle = {Proceedings of the 30th ACM Conference on Hypertext and Social Media}, keywords = {wikipedia}, pages = {143–152}, publisher = {ACM}, series = {HT '19}, title = {On the Right Track! Analysing and Predicting Navigation Success in Wikipedia}, year = 2019 }