Dr Franziska Horn

Dr Franziska Horn has been working as an AI consultant at WPS since January 2026. She completed her doctorate in machine learning at TU Berlin in 2020 and combines sound scientific expertise with many years of practical experience as a data scientist in start-ups and large companies - especially in the process industry.

Her focus is on the targeted use of artificial intelligence: AI should not be implemented as an end in itself, but to solve concrete and economically relevant problems. In doing so, it consistently thinks about ML projects from the perspective of productive use - from clean model evaluation and robust data and training pipelines through to sustainable MLOps structures. The aim is to develop solutions that do not end up as prototypes, but create long-term added value and work reliably with up-to-date data for years to come.

Publications

• A Practitioner's Guide to Machine Learning
  Franziska Horn - 14 November 2021 (modified: 02 January 2026)
• Clarity-Driven Development of Scientific Software
  Franziska Horn - leanpub.com, 16 March 2025 (modified: 25 October 2025)
• Exploring Word Usage Change with Continuously Evolving Embeddings
  Franziska Horn - In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing: System Demonstrations, pages 290-297, Online, August 2021. Association for Computational Linguistics (ACL).
• Forecasting Industrial Aging Processes with Machine Learning Methods
  Mihail Bogojeski, Simeon Sauer, Franziska Horn, Klaus-Robert Müller - Computers and Chemical Engineering, 144:107123, 2021.
• The autofeat Python Library for Automatic Feature Engineering and Selection
  Franziska Horn, Robert Pack, Michael Rieger - ECML PKDD Workshops 2019, Springer, Cham, 2020.
• Automating the search for a patent's prior art with a full text similarity search
  Lea Helmers*, Franziska Horn*, Franziska Biegler, Tim Oppermann, Klaus-Robert Müller - PLoS ONE, 14(3):e0212103, 2019.
• Predicting Pairwise Relations with Neural Similarity Encoders
  Franziska Horn, Klaus-Robert Müller - Bulletin of the Polish Academy of Sciences: Technical Sciences, 66(6):821-830, 2018
• Context encoders as a simple but powerful extension of word2vec
  Franziska Horn - In Proceedings of the 2nd Workshop on Representation Learning for NLP, pages 10-14, Vancouver, Canada, August 2017. ACL.
• "What is Relevant in a Text Document?":An Interpretable Machine Learning Approach
  Leila Arras, Franziska Horn, Gregoire Montavon, Klaus-Robert Müller, Wojciech Samek - PLoS ONE, 12(8):e0181142, 2017.
• Explaining Predictions of Non-Linear Classifiers in NLP
  Leila Arras, Franziska Horn, Gregoire Montavon, Klaus-Robert Müller, Wojciech Samek - In Proceedings of the 1st Workshop on Representation Learning for NLP, pages 1-7, Berlin, Germany, August 2016. ACL.
• Robust Artifactual Independent Component Classification for BCI Practitioners
  I. Winkler, S. Brandl, F. Horn, E. Waldburger, C. Allefeld, M. Tangermann - Journal of Neural Engineering, 11(3):035013, 2014.
• Predicting Treatment Response in Social Anxiety Disorder From Functional Magnetic Resonance Imaging
  O. Doehrmann, S. S. Ghosh, F. E. Polli, G. O. Reynolds, F. Horn, A. Keshavan, ... & J. D. Gabrieli - JAMA Psychiatry, 70(1):87-97, 2013.
• Increasing the Spectral Signal-To-Noise Ratio of Common Spatial Patterns
  Franziska Horn, Sven Dähne - Proceedings of the Fifth International Brain-Computer Interface Meeting, 2013.
• Combining Multiple EEG Features in Motor Imagery BCI
  Franziska Horn, Johannes Höhne, Sven Dähne, Benjamin Blankertz - BBCI Workshop - Advances in Neurotechnology, Berlin, Germany, 2012.