DFG Research Training Group 2153: "Energy Status Data - Informatics Methods for its Collection, Analysis and Exploitation"
Holger Trittenbach

M.Sc. Holger Trittenbach

  • Lehrstuhl
    Prof. K. Böhm

    Karlsruher Institut für Technologie
    ​Institut für Programmstrukturen und Datenorganisation
    Am Fasanengarten 5,
    76131 Karlsruhe
    GERMANY   

Research Abstract

Energy status data is typically time related and contains useful information to understand real world systems. Advanced measurement technology allows to collect data from such systems with high dimensionality and high volume. For example, in a production facility, modern smart meters allow to measure different physical quantities like voltage, frequency and harmonic distortion. With sample rates up to multiple measurements per second, these devices produce huge amounts of data. The data collected can give an indication about the behavior of machines and the quality of the electrical grid. In such a scenario, it is often of interest to detect unusual patterns which can reveal system misconfiguration or predict critical failures.

Most approaches to detect unusual patterns leave the interpretation of algorithmic results to the user. In addition, users need an advanced understanding of the detection methods to be able to adapt the algorithms to their needs. In the case of supervised learning, users must also provide the algorithm with additional ground truth information, such as the labeling of already known unusual patterns. The high dimensionality and the enormous volume of energy data makes this a challenging and time consuming task.

Focus of the doctoral thesis is to develop new approaches which utilize user feedback to advance the identification and interpretation of unusual patterns for high- dimensional data sets. There are two sides to this problem. First, the attention of a user is a scarce resource. Hence, mechanisms to request additional information from the user need to be designed carefully. Second, most methods for high-dimensional outlier detection are unsupervised and have no means to make use of labeled data. This calls for novel methods that consider both aspects. In addition, evaluation of novel approaches in this area is a challenge by itself. Not all identified anomalous patterns might be of interest to the user and the extent of interest can depend on individual preferences. This makes the quantification of interpretability and usefulness an open research task.

Research interests

  • Data Mining
  • Outlier Detection in High Dimensional Data Sets

Publications


Active Learning of SVDD Hyperparameter Values.
Trittenbach, H.; Böhm, K.; Assent, I.
2019
Active Learning of SVDD Hyperparameter Values.
Trittenbach, H.; Böhm, K.; Assent, I.
2020. 2020 IEEE 7th International Conference on Data Science and Advanced Analytics: 6-9 Ocotber 2020, Sydney, Australia. Ed.: G. Webb, 109–117, Institute of Electrical and Electronics Engineers (IEEE). doi:10.1109/DSAA49011.2020.00023
An overview and a benchmark of active learning for outlier detection with one-class classifiers.
Trittenbach, H.; Englhardt, A.; Böhm, K.
2021. Expert systems with applications, 168, Art. Nr.: 114372. doi:10.1016/j.eswa.2020.114372
Finding the sweet spot: Batch selection for one-class active learning.
Englhardt, A.; Trittenbach, H.; Vetter, D.; Böhm, K.
2020. Proceedings of the 2020 SIAM International Conference on Data Mining. Ed.: C. Demeniconi, 118–126, SIAM. doi:10.1137/1.9781611976236.14
User-Centric Active Learning for Outlier Detection. PhD dissertation.
Trittenbach, H.
2020, March 2. Karlsruher Institut für Technologie (KIT). doi:10.5445/IR/1000117443
One-class active learning for outlier detection with multiple subspaces.
Trittenbach, H.; Böhm, K.
2019. CIKM ’19 Proceedings of the 28th ACM International Conference on Information and Knowledge Management, Beijing, China, November 3-7, 2019, 811–820, Association for Computing Machinery (ACM). doi:10.1145/3357384.3357873
Understanding the effects of temporal energy-data aggregation on clustering quality.
Trittenbach, H.; Bach, J.; Böhm, K.
2019. Information technology, 61 (2-3), 111–123. doi:10.1515/itit-2019-0014
Validating one-class active learning with user studies – A prototype and open challenges.
Trittenbach, H.; Englhardt, A.; Böhm, K.
2019. IAL 2019 Interactive Adaptive Learning: Proceedings of the Workshop on Interactive Adaptive Learning co-located with European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML PKDD 2019), Würzburg, Germany, September 16th, 2019. Ed.: D. Kottke, 17–31
Data-driven crack assessment based on surface measurements.
Schulz, K.; Kreis, S.; Trittenbach, H.; Böhm, K.
2019. Engineering fracture mechanics, 218, Article no: 106552. doi:10.1016/j.engfracmech.2019.106552
The Effect of Temporal Aggregation on Battery Sizing for Peak Shaving.
Werle, D.; Warzel, D.; Bischof, S.; Koziolek, A.; Trittenbach, H.; Böhm, K.
2019. Tenth ACM International Conference on Future Energy Systems (ACM e-Energy) and its co-located workshops, Phoenix, AZ, United States, 25th - 28th of June 2019, 482–485, Association for Computing Machinery (ACM). doi:10.1145/3307772.3331023
Energy Time-Series Features for Emerging Applications on the Basis of Human-Readable Machine Descriptions.
Vollmer, M.; Englhardt, A.; Trittenbach, H.; Bielski, P.; Karrari, S.; Böhm, K.
2019. Tenth ACM International Conference on Future Energy Systems (ACM e-Energy) and its co-located workshops, Phoenix, AZ, United States, 25th - 28th of June 2019, 474–481, Association for Computing Machinery (ACM). doi:10.1145/3307772.3331022
Towards Simulation-Data Science : A Case Study on Material Failures.
Trittenbach, H.; Gauch, M.; Böhm, K.; Schulz, K.
2018. IEEE 5th International Conference on Data Science and Advanced Analytics (DSAA), Turin, Italy, 1-3 Oct. 2018, 450–459, Institute of Electrical and Electronics Engineers (IEEE). doi:10.1109/DSAA.2018.00058
An Overview and a Benchmark of Active Learning for One-Class Classification.
Trittenbach, H.; Englhardt, A.; Böhm, K.
2018. arXiv preprint 1808.04759
On the Tradeoff between Energy Data Aggregation and Clustering Quality.
Trittenbach, H.; Bach, J.; Böhm, K.
2018. 9th ACM International Conference on Future Energy Systems (ACM e-Energy), Karlsruhe, June 12-15,2018
HIPE – An energy-Status-Data set from industrial production.
Bischof, S.; Trittenbach, H.; Vollmer, M.; Werle, D.; Blank, T.; Böhm, K.
2018. 9th ACM International Conference on Future Energy Systems, e-Energy 2018; Karlsruhe; Germany; 12 June 2018 through 15 June 2018, 599–603, Association for Computing Machinery (ACM). doi:10.1145/3208903.3210278
On the tradeoff between energy data aggregation and clustering quality.
Trittenbach, H.; Bach, J.; Böhm, K.
2018. 9th ACM International Conference on Future Energy Systems, e-Energy 2018; Karlsruhe; Germany; 12 June 2018 through 15 June 2018, 399–401, Association for Computing Machinery (ACM). doi:10.1145/3208903.3212038
Dimension-based subspace search for outlier detection.
Trittenbach, H.; Böhm, K.
2018. International Journal of Data Science and Analytics. doi:10.1007/s41060-018-0137-7
HIPE -- An Energy-Status-Data Set from Industrial Production.
Bischof, S.; Trittenbach, H.; Vollmer, M.; Werle, D.; Blank, T.; Böhm, K.
2018. International Workshop on Energy Data and Analytics (EDA 2018), Karlsruhe, Germany, June 12, 2018
Towards Simulation-Data Science : A Case Study on Material Failures.
Trittenbach, H.; Gauch, M.; Böhm, K.; Schulz, K.
2018. Karlsruher Institut für Technologie (KIT). doi:10.5445/IR/1000079420