Data Scientists spend a significant amount of time on building features which are certain characteristics or patterns in the inspected data. This is especially true for timely annotated data such as time series and sequences. Now, the python package tsfresh frees your time spend on building features by extracting them automatically from time series.
|Journal Paper||Maschine Learning||2018||
Maximilian Christ, Nils Braun, Julius Neuffer, Andreas W. Kempa-Liehr
"Time Series FeatuRe Extraction on basis of Scalable Hypothesis tests (tsfresh – A Python package)"
Journal Neurocomputing, https://www.sciencedirect.com/science/article/pii/S0925231218304843
|Conference Paper||Service Science||2017||
Maximilian Christ, Julius Neuffer, Andreas W. Kempa-Liehr
"On the Implicit Cost Structure of Service Levels from the Perspective of the Service Consumer"
Proceedings of the The 7th International Conference on Cloud Computing and Services Science (Closer’17)
|Conference Paper||Computer Science||2016||Maximilian Christ, Julian Krumeich, Andreas W. Kempa-Liehr:
"Integrating Predictive Analytics into Complex Event Processing by Using Conditional Density Estimations."
Enterprise Distributed Object Computing Workshop (EDOCW), 2016 IEEE 20th International. IEEE, 2016.
|Conference Paper||Computer Science||2016|| Michael Falkenthal, Uwe Breitenbücher, Maximilian Christ, Christian Endres, Andreas W. Kempa-Liehr, Frank Leymann, Michael Zimmermann:
"Towards Function and Data Shipping in Manufacturing Environments: How Cloud Technologies leverage the 4th Industrial Revolution."
Proceedings of the 10th Advanced Summer School on Service Oriented Computing
|Conference Poster||Data Science||2016|| Maximilian Christ, Frank Kienle, Andreas W. Kempa-Liehr
"Time Series Analysis in Industrial Applications"
In: Workshop on Extreme Value and Time Series Analysis (Extremes 2016)
|Conference Paper||Machine Learning||2016||
Maximilian Christ, Andreas W. Kempa-Liehr, Michael Feindt
"Distributed and parallel time series feature extraction for industrial big data applications."
Workshop on Learning on Big Data, Asian Conference on Machine Learning (ACML 2016)
ArXiv e-print 1610.07717, https://arxiv.org/abs/1610.07717
|Conference Paper||Computer Science||2016||
Michael Falkenthal, Uwe Breitenbücher, Kálmán Képes, Frank Leymann, Michael Zimmermann, Maximilian Christ, Julius Neuffer, Nils Braun, Andreas W. Kempa-Liehr
"OpenTOSCA for the 4th Industrial Revolution: Automating the Provisioning of Analytics Tools based on Apache Flink."
Proceedings of the 6th International Conference on the Internet of Things (IoT'16 )
I am a passionate Mathematician who pursues a Ph.D. besides his business work. As a Data Scientist, my interests and experiences in Programming, Classical Statistics, Distributed Systems and Machine Learning help me to realize measurable business value.
As a Freelance Data Scientist / Data Engineer I have consulted clients from the Internet of Things (IoT) and Finance environments. I focused on developing systems that solved dynamic problems, which involved the analysis of time series with help of statistics and machine learning. In finished projects I have implemented an automated annotating of sensory time series or developed a classifier for an intraday trading strategy.October 2017
Consulting clients about Predictive Applications in the industrial sector. Finished projects investigated the statistical and technical feasibility of data driven decisions for Industrial Applications. Such can be used for predictive maintenance as well as for the optimization of discrete and continuous production of simple (e.g. steel billets) and complex goods (e.g. electronic control units). Further participating in the BMBF and BMWi funded research projects "iPRODICT" and "Sepia.Pro".March 2015
Testing the calculation of premiums for life insurances and fund based saving plans in the company wide inventory management system by designing and implementing simulators. The simulators were written in PERL and VBA and the inventory management system itself was written in C.Oktober 2014
Employing of statistical techniques to find fraudulent behavior patterns in non homogenous Big Data sources such as ERP data or textual sources. Resulting analytic components were written in Java.Juli 2014
Specialization in Functional Analysis and Statistics. Master thesis extended Logic Regression by a feature learning mechanism.April 2013
Two abroad semester in the Erasmus program. Specialization in Machine Learning and Finance. Sí, hablo español ;).August 2013
Specialization in Statistics.April 2010