Contact:
| Address: |
Postal Address Humboldt-Universität zu Berlin Department of Computer Science Unter den Linden 6 10099 Berlin Visiting Address |
| Phone: |
030-2093-3914 |
| EMail: |
patrick.schaefer
hu-berlin.de
|
Teaching:
Source codes:
- Java: Scalable Time Series Data Analytics using SFA, BOSS, BOSS VS and WEASEL
- Python: Motiflets.
- Python: WEASEL v1 and v2, SFA, BOSS.
- Sktime - A unified framework for machine learning with time series.
Selected Publications
- A. Ermshaus, P. Schäfer, U. Leser, ClaSP: parameter-free time series segmentation, DAMI, 2023(Webpage, Paper, Code)
- P. Schäfer, U. Leser, WEASEL 2.0 - A Random Dilated Dictionary Transform for Fast, Accurate and Memory Constrained Time Series Classification, ArXiv Preprint 2023, (Paper, Webpage/Code)
- A. Ermshaus, P. Schäfer, U. Leser, Window Size Selection In Unsupervised Time Series Analytics: A Review and Benchmark, AALTD @ ECML, 2022, (Paper, Webpage/Code)
- P. Schäfer, U. Leser, Motiflets - Fast and Accurate Detection of Motifs in Time Series, VLDB 2023. (Webpage, Paper, Code)
- P. Schäfer, A. Ermshaus, U. Leser, ClaSP - Time Series Segmentation, CIKM, Virtual, 2021(Webpage, Paper, Code)
- P. Schäfer, U. Leser, TEASER: Early and Accurate Time Series Classification, ECML / PKDD, Virtual, 2020. (Webpage, Paper, Code)
- S. Sprenger, P. Schäfer, Ulf Leser, BB-Tree: A practical and efficient main-memory index structure for multidimensional workloads. Int. Conf. on Extending Database Technologies (EDBT) Lisbon, Portugal, 2019. (Paper)
- S. Sprenger, P. Schäfer, Ulf Leser, BB-Tree: A main-memory index structure for multidimensional range queries (short paper). Int. Conf. on Data Engineering, Macau, China, 2019.
- P. Schäfer, Ulf Leser, Multivariate Time Series Classification with WEASEL+MUSE, 3rd ECML/PKDD Workshop on AALTD, 2018. (Paper, Code)
- S. Sprenger, P. Schäfer, Ulf Leser, Multidimensional Range Queries on Modern Hardware, Prepint. (Paper)
- P. Schäfer, Dirk Pflugmacher, Patrick Hostert, Ulf Leser, Classifying land cover from satellite images using time series analytics, Workshop Proceedings of the EDBT/ICDT 2018. (Paper)
- P. Schäfer, Ulf Leser, Multivariate Time Series Classification with WEASEL+MUSE, Prepint. (Webpage, Paper, Code)
- P. Schäfer, Ulf Leser, Fast and Accurate Time Series Classification with WEASEL, CIKM 2017. (Webpage, Paper, Code)
- P. Schäfer, Ulf Leser, Benchmarking Univariate Time Series Classifiers, BTW 2017 (Paper)
- P. Schäfer, Scalable Time Series Classification, ECML/PKDD 2016, DMKD 2016 (Webpage, Paper, Keynote, PPT, Code)
- P. Schäfer, Scalable Time Series Classification for Data Analytics, Dissertation 2015, (Dissertation, Keynote, PPT)
- P. Schäfer, The BOSS is concerned with time series classification in the presence of noise, DMKD 2014, (Webpage, Paper, Code)
- P. Schäfer, Experiencing the Shotgun Distance for Time Series Analysis, Transactions MLDM 2014, (Webpage, Paper, Code)
- I. Potamitis, P. Schäfer, On classifying insects from their wing-beat: New results, Ecology and acoustics: emergent properties from community to landscape, Paris, France 2014.
- P. Schäfer, Towards Time Series Classification without Human Preprocessing, MLDM 2014, (Webpage, Paper, Code)
- P. Schäfer, Shooting Audio Recordings of Insects with SFA, AmiBio Workshop 2013.
- P. Schäfer, S. Dressler, SFA : a symbolic fourier approximation and index for similarity search in high dimensional datasets, EDBT 2012, (Webpage, Paper, Code)
Research Interests:
- Big Data Analytics
- Time Series
- Information Retrieval
- NoSQL
Reviewer & Program Committee:
- ECML since '21
- AAAI since '20
- IJCAI '20
- ADBIS '20
- AALTD since '16 - "Advanced Analytics and Learning on Temporal Data"
- LDWA '19 - "Large-Scale Data Management and Processing - Applications in Research and Industry"
- IEEE/CAA '19 - Special Issue on Time Series Classification
- TKDE, since '17
- DAMI, since '16
- IEEE Cybernetics, since '18
Current Projects:
Past Projects:
- sktime (Former Core developer)
- MoSGrid
- Harness
- Contrail
- XtreemFS