To top

Vita

In the beginning of 2015, I graduated in the Master's Program "Language Science and Technology" in Saarbrücken.

Before, I had completed my Bachelor's degree at Heinrich-Heine Universität Düsseldorf as a student of "Information Science and Language Technology".

During the half-year period between the two programs, I had earned Credit Points in the international Master's Program on Artificial Intelligence in Maastricht - but ultimately decided to specialize further in the field of Language Technology / Computational Linguistics.

At GESIS Cologne, I can build upon my background in Information Science and broaden my knowledge in the field of Language Technology. Here, I am working in the OUTCITE project to extract and link references from PDF documents.
To top

Service

I am currently involved in the OUTCITE project, where we try to link references extracted from fulltext PDF's to authoritative URL addresses. My current contributions are in evaluation techniques for extracted references and in implementing various information retrieval approaches to the task of finding good URL's matching these references. To top

Research

My research focus is on entity resolution such as (progressive) author disambiguation, (hierarchical) institution resolution and billion-scale duplicate detection. I am in the process of completing a PhD thesis that addresses these application scenarios in a common (formal) framework.
To top

Publications

Publication

To top

Journal article

Backes, Tobias, and Stefan Dietze. 2022. "Lattice-based progressive author disambiguation." Information Systems 109 (November 2022): 102056. doi: https://doi.org/10.1016/j.is.2022.102056.

Backes, Tobias, Daniel Hienert, and Stefan Dietze. 2022. "Towards Hierarchical Affiliation Resolution: Framework, Baselines, Dataset." International Journal on Digital Libraries 23 (3): 267-288. doi: https://doi.org/10.1007/s00799-022-00326-1.

To top

Contribution to edited volume

Backes, Tobias. 2018. "Effective unsupervised author disambiguation with relative frequencies." In Proceedings of the 16th ACM/IEEE Joint Conference on Digital Libraries (JCDL'18), edited by Jiangping Chen, Marcos André Gonçalves, and Jeff M. Allen, doi: https://doi.org/10.1145/3197026.3197036.

Backes, Tobias. 2018. "The Impact of Name-Matching and Blocking on Author Disambiguation." In CIKM '18 : The 27th ACM International Conference on Information and Knowledge Management Torino, Italy — October 22 - 26, 2018, 803-812. New York: ACM. doi: https://doi.org/10.1145/3269206.3271699.

Lecture

To top

Conference contribution

Backes, Tobias. 2018. "Effective Unsupervised Author Disambiguation with Relative Frequencies." Joint Conference on Digital Libraries (JCDL) 2018, 2018-06-05.

Backes, Tobias. 2018. "The Impact of Name-Matching and Blocking on Author Disambiguation." 27th ACM International Conference on Information and Knowledge Management, 2018-10-24.