Prof. Dr. Emmanuel Müller

Arvind Kumar S

PhD Student in collaboration with Bosch GmbH

Chair for Knowledge Discovery and Data Mining


Contact Information

email: arvindkumar.shekar(at)de.bosch.com

office: E Block, X-0Y.Y, where

X= Multiplicative Identity
= Smallest Prime Number
(Kindly solve this to find my office)

Research Focus

  • Methods: Feature selection, Regression, Classification, Time series analysis, Multi-variate correlation measures, Ordinal Pattern
  • Analysis Data: Automotive components
  • Applications: Predictive diagnostics, root cause analysis, failure prediction

Academic and Professional experience

  • [2015-now] PhD Student at HPI in collaboration with Bosch Diesel Systems, advisor: Prof. Dr. Emmanuel Mueller
  • [2013-2015] MSc in Electronics Engineering, Hochschule Bremen, advisor: Prof. Dr. Dieter Kraus
  • [2011-2013] Software Engineer at Robert Bosch Engineering and Business Solutions India Limited
  • [2007-2011] Bachelors in Mechatronics Engineering, KCT, India

Feature selection on high dimensional datasets

Automobile control units monitor several hundreds of signals. In addition to the high dimensionality, these signals exhibit strong interactions between them. Hence, the failure of one sensor can lead to disturbances in multiple signals. This project involves automatic selection of the sensor signals that are strongly influenced by a particular failure.

By performing such cause and effect analysis in a automated framework, the effort and time spent by the engineers for manual analysis is greatly reduced.


Are you creative? Do you like to solve new industrial challenges? We provide exciting master theses in this domain. For further details feel free to write me an e-mail.


1. Arvind Kumar Shekar, Patricia Iglesias Sanchez and Emmanuel Mueller. "Diverse Selection of Feature Subsets for Ensemble Regression" at 19th International Conference on Big Data Analytics and Knowledge Discovery - DaWaK 2017 in France.

2. Arvind Kumar Shekar, Tom Bocklisch, Patricia Iglesias Sanchez, Christoph Nikolas Straehle and Emmanuel Mueller. "Multi-Feature Interactions and Redundancy for Feature Ranking in Mixed Data"  at ECML-PKDD 2017 in Macedonia.

3. Louis Kirsch, Niklas Riekenbrauck, Daniel Thevessen, Marcus Pappik, Axel Stebner, Julius Kunze, Alexander Meissner, Arvind Kumar Shekar and Emmanuel Mueller. "Framework for Exploring and Understanding Multivariate Correlations" at ECML-PKDD 2017 in Macedonia.

4. Arvind Kumar Shekar, Cláudio Rebelo de Sá, Hugo Ferreira and Carlos Soares . "Building robust prediction models for defective sensor data using Artificial Neural Networks " arXiv preprint arXiv:1804.05544 (2018).

Link to my: [DBLP] [Google Scholar]


Teaching Experience

  • [2016] Supervision of Master Thesis
    "Including multi-feature interactions and redundancy for feature ranking in mixed datasets"
    by Tom Bocklisch
  • [2016-2017] Supervision of Bachelor-Project "Enhancing Machine Learning for Predictive Diagnostics"

Other Activities

External Reviewer:

  • VLDB(2017)
  • KDD (2017)
  • CIKM (2016/2017)

Machine Learning Session Chair at The 19th International Conference on Big Data Analytics and Knowledge Discovery - DaWaK 2017 in France


Patent in "Verfahren zur Steuerung des Zugriffs in einem Fahrzeug" (2626 / CHE / 2013) bei Robert Bosch India Limited eingereicht