Prof. Dr. Felix Naumann

Dr. Michael Loster

Postdoctoral Researcher

für Softwaresystemtechnik
Prof.-Dr.-Helmert-Straße 2-3
D-14482 Potsdam

Phone: +49 331 5509 286
Fax: +49 331 5509 287
Room: F-2.08
Email: Michael Loster
Twitter: @miclost

Homepage: loster.io
Research: GoogleScholar, DBLPResearchGate, GitHub


Research Interests

  • Knowledge Engineering
  • Text Mining
  • Machine Learning
  • Deep Learning
  • Natural Language Processing
  • Opinion Mining
  • Data Mining


  • COLT: A few-shot knowledge validation approach using rules
  • SNNDedupe: An neural approach for entity resolution leveraging siamese neural networks and knowledge transfer
  • CurEx: A system for extracting, curating, and exploring domain-specific knowledge graphs


Bachelor Projects:

  • Contagion: Analytical Tools for Semantic Company Networks (2015/2016)
  • Ingestion: Data Acquisition and Analysis for Semantic Enterprise Networks (2016/2017)

Master Thesis:

  • From Text to Facts: Relation Extraction on German Company Websites (Tanja Bergmann, 2016)
  • German Organization Name Part Classification (Manuel Hegner, 2016)

Bachelor Thesis:

  • Evaluation of Entity Linking Models on Business Data (Jan Ehmueller, 2017)
  • Analysis and Simplification of Business Graphs (Milan Gruner, 2017)
  • Distributed Business Relations in Apache Cassandra (Nils Strelow, 2017)
  • Transforming Company Network Information out of multiple heterogenous Sources into a Graph Database (Marvin Luca Gorecki, 2016)
  • Name Matching German Companies (Willi Raschkowski, 2016)
  • Visualization of Enterprise Networks (Benjamin Feldmann, 2016)
  • Building a Semantic Company Graph using Knowledge Bases (Fabian Windheuser, 2016)
  • Generating Aliases to Improve Company Recognition in Text (Alexander Immer, 2016)


  • Loster, Michael, Davide Mottin, Paolo Papotti, Felix Naumann, Jan Ehmueller, and Benjamin Feldmann. ‘Few-Shot Knowledge Validation Using Rules’. In Proceedings of the Web Conference, 2021.
  • Loster, Michael, Ioannis Koumarelas, and Felix Naumann. ‘Knowledge Transfer for Entity Resolution With Siamese Neural Networks’. Journal of Data and Information Quality 13, no. 1 (January 2021). https://doi.org/10.1145/3410157.
  • CurEx: A System for Extra... - Download
    Loster, Michael, Felix Naumann, Jan Ehmueller, and Benjamin Feldmann. ‘CurEx: A System for Extracting, Curating, and Exploring Domain-Specific Knowledge Graphs from Text’. In Proceedings of the ACM International Conference on Information and Knowledge Management, 1883–1886. ACM, 2018. https://doi.org/10.1145/3269206.3269229.
  • Dissecting Company Names ... - Download
    Loster, Michael, Manuel Hegner, Felix Naumann, and Ulf Leser. ‘Dissecting Company Names Using Sequence Labeling’. In Proceedings of the Conference "Lernen, Wissen, Daten, Analysen", 2191:227–238. CEUR Workshop Proceedings, 2018. http://ceur-ws.org/Vol-2191/paper27.pdf.
  • The Challenges of Creatin... - Download
    Loster, Michael, Tim Repke, Ralf Krestel, Felix Naumann, Jan Ehmueller, Benjamin Feldmann, and Oliver Maspfuhl. ‘The Challenges of Creating, Maintaining and Exploring Graphs of Financial Entities’. In Proceedings of the Fourth International Workshop on Data Science for Macro-Modeling (DSMM 2018). ACM, 2018. https://doi.org/10.1145/3220547.3220553.
  • Uncovering Business Relat... - Download
    Zuo, Zhe, Michael Loster, Ralf Krestel, and Felix Naumann. ‘Uncovering Business Relationships: Context-Sensitive Relationship Extraction for Difficult Relationship Types’. In Proceedings of the Conference "Lernen, Wissen, Daten, Analysen" (LWDA), 2017.
  • Improving Company Recogni... - Download
    Loster, Michael, Zhe Zuo, Felix Naumann, Oliver Maspfuhl, and Dirk Thomas. ‘Improving Company Recognition from Unstructured Text by Using Dictionaries’. In Proceedings of the International Conference on Extending Database Technology, 610–619, 2017. https://doi.org/10.5441/002/edbt.2017.82.
  • Comparing Features for Ra... - Download
    Repke, Tim, Michael Loster, and Ralf Krestel. ‘Comparing Features for Ranking Relationships Between Financial Entities Based on Text’. In Proceedings of the 3rd International Workshop on Data Science for Macro--Modeling With Financial and Economic Datasets, 12:1–12:2. DSMM’17. New York, NY, USA: ACM, 2017. http://doi.acm.org/10.1145/3077240.3077252.
  • Combination of Rule-based... - Download
    Samiei, Ahmad, Ioannis Koumarelas, Michael Loster, and Felix Naumann. ‘Combination of Rule-Based and Textual Similarity Approaches to Match Financial Entities’. In Data Science for Macro-Modeling With Financial and Economic Datasets (DSMM). ACM, 2016. http://dl.acm.org/citation.cfm?id=2951905.