Jump to ContentJump to Main Navigation
Show Summary Details
More options …

at - Automatisierungstechnik

Methoden und Anwendungen der Steuerungs-, Regelungs- und Informationstechnik

[AT - Automation Technology: Methods and Applications of Control, Regulation, and Information Technology

Editor-in-Chief: Jumar, Ulrich

IMPACT FACTOR 2018: 0.500

CiteScore 2018: 0.60

SCImago Journal Rank (SJR) 2018: 0.211
Source Normalized Impact per Paper (SNIP) 2018: 0.532

See all formats and pricing
More options …
Volume 67, Issue 10


Feature fusion to increase the robustness of machine learners in industrial environments

Merkmalsfusion zur Erhöhung der Robustheit von maschinellen Lernern in industriellen Umgebungen

M. Sc. Christoph-Alexander Holst
  • Corresponding author
  • inIT – Institute Industrial IT, Technische Hochschule Ostwestfalen-Lippe, Campusallee 6, 32657 Lemgo, Germany
  • Email
  • Other articles by this author:
  • De Gruyter OnlineGoogle Scholar
/ Prof. Dr.-Ing. Volker Lohweg
Published Online: 2019-09-27 | DOI: https://doi.org/10.1515/auto-2019-0028


Industrial applications put special demands on machine learning algorithms. Noisy data, outliers, and sensor faults present an immense challenge for learners. A considerable part of machine learning research focuses on the selection of relevant, non-redundant features. This contribution details an approach to group and fuse redundant features prior to learning and classification. Features are grouped relying on a correlation-based redundancy measure. The fusion of features is guided by determining the majority observation based on possibility distributions. Furthermore, this paper studies the effects of feature fusion on the robustness and performance of classification with a focus on industrial applications. The approach is statistically evaluated on public datasets in comparison to classification on selected features only.


Industrielle Anwendungen stellen besondere Anforderungen an maschinelle Lernalgorithmen. Verrauschte Daten, Ausreißer und Sensorfehler sind eine große Herausforderung für Lerner. Ein erheblicher Teil der Forschung im Bereich maschinelles Lernen konzentriert sich auf die Auswahl relevanter, nicht redundanter Merkmale. Dieser Beitrag beschreibt einen Ansatz zur Gruppierung und Fusion redundanter Merkmale, die vor dem Lernen und Klassifizieren ausgeführt werden. Die Gruppierung der Merkmale basiert auf einer korrelationsbasierten Redundanzmessung, während die Fusionierung von Merkmalen auf der Bestimmung einer über Möglichkeitsverteilungen ermittelten Mehrheitsbeobachtung basiert. Darüber hinaus untersucht dieser Beitrag die Auswirkungen der Merkmalsfusion auf die Robustheit und Leistungsfähigkeit der Klassifizierung. Der Ansatz wird anhand öffentlicher Datensätze im Vergleich zur Klassifizierung auf ausgewählten Merkmalen statistisch ausgewertet.

Keywords: robust machine learning; feature grouping; feature fusion; information fusion

Schlagwörter: Robustes Maschinelles Lernen; Merkmalsgruppierung; Merkmalsfusion; Informationsfusion


  • 1.

    M. A. Aizerman, E. M. Braverman and L. I. Rozonoer. Theoretical foundations of the potential function method in pattern recognition learning. Automation and Remote Control, 25:821–837, 1964.Google Scholar

  • 2.

    E. Alpaydın. Introduction to Machine Learning. The MIT Press, Cambridge, 2nd edition, 2010.Google Scholar

  • 3.

    B. M. Ayyub and G. J. Klir. Uncertainty Modeling and Analysis in Engineering and the Sciences. Chapman & Hall/CRC, Boca Raton, FL, 2006.Google Scholar

  • 4.

    J. Beyerer, J. Jasperneite and O. Sauer. Industrie 4.0. at – Automatisierungstechnik, 63(10), 2015.Google Scholar

  • 5.

    F. Bocklisch and D. Hausmann. Multidimensional fuzzy pattern classifier sequences for medical diagnostic reasoning. Applied Soft Computing, 66:297–310, 2018.Web of ScienceCrossrefGoogle Scholar

  • 6.

    S. F. Bocklisch. Prozeßanalyse mit unscharfen Verfahren. Verlag Technik, Berlin, 1st edition, 1987.Google Scholar

  • 7.

    L. Breiman. Bagging predictors. Machine Learning, 24(2):123–140, 1996.CrossrefGoogle Scholar

  • 8.

    L. Breiman. Classification and Regression Trees. Routledge, New York, 2017.Google Scholar

  • 9.

    D. Dheeru and E. Karra Taniskidou. UCI Machine Learning Repository, 2017.

  • 10.

    T. G. Dietterich. Approximate statistical tests for comparing supervised classification learning algorithms. Neural Computation, 10(7):1895–1923, 1998.CrossrefGoogle Scholar

  • 11.

    A. Diez-Olivan, J. Del Ser, D. Galar and B. Sierra. Data fusion and machine learning for industrial prognosis: Trends and perspectives towards Industry 4.0. Information Fusion, 50:92–111, 2019.CrossrefWeb of ScienceGoogle Scholar

  • 12.

    H. Dörksen and V. Lohweg. Combinatorial refinement of feature weighting for linear classification. In Proceedings of the 2014 IEEE Emerging Technology and Factory Automation (ETFA), pages 1–7, 2014.Google Scholar

  • 13.

    D. Dubois, L. Foulloy, G. Mauris and H. Prade. Probability-possibility transformations, triangular fuzzy sets, and probabilistic inequalities. Reliable Computing, 10(4):273–297, 2004.CrossrefGoogle Scholar

  • 14.

    J.-F. Ehlenbröker, U. Mönks and V. Lohweg. Sensor defect detection in multisensor information fusion. Journal of Sensors and Sensor Systems, 5(2):337–353, 2016.CrossrefWeb of ScienceGoogle Scholar

  • 15.

    W. Elmenreich. An Introduction to Sensor Fusion, 2002.

  • 16.

    A. Fritze, U. Mönks, C.-A. Holst and V. Lohweg. An approach to automated fusion system design and adaptation. Sensors, 17(3):601, 2017.Web of ScienceCrossrefGoogle Scholar

  • 17.

    S. Glock, K. Voth, J. Schaede and V. Lohweg. A framework for possibilistic multi-source data fusion with monitoring of sensor reliability. In World Conference on Soft Computing, 2011.Google Scholar

  • 18.

    I. Guyon, S. R. Gunn, M. Nikravesh and L. A. Zadeh. Feature extraction: Foundations and applications, volume 207 of Studies in Fuzziness and Soft Computing. Springer, Berlin Heidelberg, 2006.Google Scholar

  • 19.

    D. L. Hall, J. Llinas and M. E. Liggins, editors. Handbook of Multisensor Data Fusion: Theory and Practice. The Electrical Engineering and Applied Signal Processing Series. CRC Press, Boca Raton, FL, 2nd edition, 2009.Google Scholar

  • 20.

    T. Hastie, R. Tibshirani, D. Botstein and P. Brown. Supervised harvesting of expression trees. Genome Biology, 2(1):research0003.1, 2001.Google Scholar

  • 21.

    N. Helwig, E. Pignanelli and A. Schütze. Detecting and compensating sensor faults in a hydraulic condition monitoring system. In SENSOR 2015, pages 641–646, Nürnberg, 2015. AMA Service GmbH.Google Scholar

  • 22.

    A.-J. Hempel. Netzorientierte Fuzzy-Pattern-Klassifikation nichtkonvexer Objektmengenmorphologien. Doctoral thesis, Technische Universität Chemnitz, Chemnitz, 2011.Google Scholar

  • 23.

    C.-A. Holst and V. Lohweg. Improving majority-guided fuzzy information fusion for Industry 4.0 condition monitoring. In 2019 22nd International Conference on Information Fusion (FUSION). IEEE, 2019.Google Scholar

  • 24.

    Z. Hu and S. Mahadevan. Uncertainty quantification in prediction of material properties during additive manufacturing. Scripta Materialia, 135:135–140, 2017.CrossrefWeb of ScienceGoogle Scholar

  • 25.

    E. Hüllermeier. Fuzzy methods in machine learning and data mining: Status and prospects. Fuzzy Sets and Systems, 156(3):387–406, 2005.CrossrefGoogle Scholar

  • 26.

    W. Jiang, C. Xie, M. Zhuang, Y. Shou and Y. Tang. Sensor data fusion with Z-numbers and its application in fault diagnosis. Sensors, 16(9), 2016.Web of ScienceGoogle Scholar

  • 27.

    W. Jiang, M. Zhuang and C. Xie. A reliability-based method to sensor data fusion. Sensors, 17(7), 2017.Web of ScienceGoogle Scholar

  • 28.

    M. Krüger. Gradual vs. binary conflicts in Bayesian networks applied to sensor failure detection. In 2015 18th International Conference on Information Fusion (Fusion), pages 66–73, 2015.Google Scholar

  • 29.

    P. Larrañaga, A. Ogbechie, J. Diaz-Rozo, D. Atienza Alonso, C. Bielza and C. Puerto-Santana. Industrial Applications of Machine Learning. Data Mining and Knowledge Series. CRC Press, Boca Raton, Florida, 2019.Google Scholar

  • 30.

    H. Li, H.-Z. Huang, Y.-F. Li, J. Zhou and J. Mi. Physics of failure-based reliability prediction of turbine blades using multi-source information fusion. Applied Soft Computing, 72:624–635, 2018.CrossrefWeb of ScienceGoogle Scholar

  • 31.

    J. Li, K. Cheng, S. Wang, F. Morstatter, R. P. Trevino, J. Tang and H. Liu. Feature selection: A data perspective. ACM Computing Surveys, 50(6):1–45, 2018.Web of ScienceGoogle Scholar

  • 32.

    V. Lohweg, C. Diederichs and D. Müller. Algorithms for hardware-based pattern recognition. EURASIP Journal on Applied Signal Processing, 2004(12):1912–1920, 2004.Google Scholar

  • 33.

    R. C. Luo and M. G. Kay. Data fusion and sensor integration: State-of-the-art 1990s. In M. A. Abidi and R. C. Gonzalez, editors, Data Fusion in Robotics and Machine Intelligence, pages 7–136. Acad. Press, San Francisco, CA, USA, 1992.Google Scholar

  • 34.

    R. Maclin and D. Opitz. An empirical evaluation of bagging and boosting. In Proceedings of the Fourteenth National Conference on Artificial Intelligence and Ninth Conference on Innovative Applications of Artificial Intelligence, AAAI’97/IAAI’97, pages 546–551. AAAI Press, 1997.Google Scholar

  • 35.

    G. Mauris, V. Lasserre and L. Foulloy. Fuzzy modeling of measurement data acquired from physical sensors. IEEE Transactions on Instrumentation and Measurement, 49(6):1201–1205, 2000.CrossrefGoogle Scholar

  • 36.

    U. Mönks. Information Fusion Under Consideration of Conflicting Input Signals. Technologies for Intelligent Automation. Springer, Berlin, Heidelberg, 2017.Google Scholar

  • 37.

    U. Mönks, D. Petker and V. Lohweg. Fuzzy-Pattern-Classifier training with small data sets. In E. Hüllermeier, R. Kruse, and F. Hoffmann, editors, Information Processing and Management of Uncertainty in Knowledge-Based Systems. Theory and Methods, pages 426–435, Berlin, Heidelberg, 2010. Springer Berlin Heidelberg.Google Scholar

  • 38.

    M. Y. Park, T. Hastie and R. Tibshirani. Averaged gene expressions for regression. Biostatistics, 8(2):212–227, 2007.Web of ScienceCrossrefGoogle Scholar

  • 39.

    H. Peng, F. Long and C. Ding. Feature selection based on mutual information: Criteria of max-dependency, max-relevance, and min-redundancy. IEEE Transactions on Pattern Analysis and Machine Intelligence, 27(8):1226–1238, 2005.CrossrefGoogle Scholar

  • 40.

    V. Ricquebourg, L. Delahoche, B. Marhic, M. Delafosse, A.-M. Jolly-Desodt and D. Menga. Anomalies recognition in a context aware architecture based on TBM approach. In 2008 11th International Conference on Information Fusion, pages 1–8, 2008.Google Scholar

  • 41.

    H. Rinne. Taschenbuch der Statistik. Wissenschaftlicher Verlag Harri Deutsch GmbH, Frankfurt am Main, 4th edition, 2008.Google Scholar

  • 42.

    F. Shi, X. Su, H. Qian, N. Yang and W. Han. Research on the fusion of dependent evidence based on rank correlation coefficient. Sensors, 17(10), 2017.Web of ScienceGoogle Scholar

  • 43.

    M. Sokolova and G. Lapalme. A systematic analysis of performance measures for classification tasks. Information Processing & Management, 45(4):427–437, 2009.Web of ScienceCrossrefGoogle Scholar

  • 44.

    A. Vergara, S. Vembu, T. Ayhan, M. A. Ryan, M. L. Homer and R. Huerta. Chemical gas sensor drift compensation using classifier ensembles. Sensors and Actuators B: Chemical, 166–167:320–329, 2012.Web of ScienceGoogle Scholar

  • 45.

    K. Voth, S. Glock, U. Mönks, V. Lohweg and T. Türke. Multi-sensory machine diagnosis on security printing machines with two-layer conflict solving. In SENSOR+TEST Conference 2011, pages 686–691, Wunstorf, 2011. AMA Service GmbH.Google Scholar

  • 46.

    L. A. Zadeh. Fuzzy sets. Information and Control, 8(3):338–353, 1965.CrossrefGoogle Scholar

About the article

M. Sc. Christoph-Alexander Holst

Christoph-Alexander Holst received his master’s degree in information technology from the Technische Hochschule Ostwestfalen-Lippe, Germany. He is working towards his doctoral degree in cooperation with the Computer Engineering Group at the Brandenburg University of Technology Cottbus-Senftenberg. His main research topics are information fusion, fusion system design, and sensor orchestration.

Prof. Dr.-Ing. Volker Lohweg

Volker Lohweg is director of the inIT – Institute Industrial IT and head of the research group Discrete Systems. The research group’s working area is dedicated to cognitive systems in automation especially information fusion in the context of intelligent technical systems. He is active in SPIE and IEEE as a reviewer in the field of image processing and data analysis. His actual interests are sensory conflict modelling and multi-scale signal analysis.

Received: 2019-02-28

Accepted: 2019-08-26

Published Online: 2019-09-27

Published in Print: 2019-10-25

Funding Source: Bundesministerium für Bildung und Forschung

Award identifier / Grant number: 01IS18041D

This work was partly funded by the German Federal Ministry of Education and Research within the project ITS.ML, Grant number 01IS18041D.

Citation Information: at - Automatisierungstechnik, Volume 67, Issue 10, Pages 853–865, ISSN (Online) 2196-677X, ISSN (Print) 0178-2312, DOI: https://doi.org/10.1515/auto-2019-0028.

Export Citation

© 2019 Walter de Gruyter GmbH, Berlin/Boston.Get Permission

Comments (0)

Please log in or register to comment.
Log in