Handling Missing Syndromes



The diagnosis accuracy of reasoning-based diagnosis engine may be significantly reduced when the repair logs are fragmented and some errors, or syndromes, are not available during diagnosis. Since root-cause isolation for a failing board relies on reasoning based on syndromes, any information loss (e.g., missing syndromes) during the extraction of a diagnosis log may lead to ambiguous repair suggestions. In this chapter, we propose a board-level diagnosis system with the feature of handling missing syndromes using the method of imputation. The syndromes from a faulty-board log are analyzed and imputed with appropriate values in a preprocessing engine before root-cause isolation. We utilize several imputation methods and compare them in terms of their effectiveness in handling missing syndromes.


Handling missing syndromes board-level Fault diagnosis Functional failure Missing syndromes Manufacturing Machine learning Feature selection Production 


  1. 1.
    Bushnell M, Agrawal V (2000) Essentials of electronic testing for digital, memory, and mixed-signal VLSI circuits. Springer, New YorkGoogle Scholar
  2. 2.
    O’Farrill C, Moakil-Chbany M, Eklow B (2005) Optimized reasoning-based diagnosis for non-random, board-level, production defects. In: Proceedings IEEE international test conference (ITC), pp 173–179Google Scholar
  3. 3.
    Ye F, Zhang Z, Chakrabarty K, Gu X (2013) Board-level functional fault diagnosis using artificial neural networks, support-vector machines, and weighted-majority voting. IEEE Trans Comput-Aided Des Integr Circuits Syst (TCAD) 32(5):723–736CrossRefGoogle Scholar
  4. 4.
    Zhang Z, Gu X, Xie Y, Wang Z, Chakrabarty K (2012) Diagnostic system based on support-vector machines for board-level functional diagnosis. In: Proceedings IEEE European test symposium (ETS), pp 1–6Google Scholar
  5. 5.
    Ye F, Zhang Z, Chakrabarty K, Gu X (2012) Adaptive board-level functional fault diagnosis using decision trees. In: Proceedings IEEE Asian test symposium (ATS), pp 202–207Google Scholar
  6. 6.
    Vo T, Wang Z, Eaton T, Ghosh P, Li H, Lee Y, Wang W, Jun H, Fang R, Singletary D, Gu X (2006) Design for board and system level structural test and diagnosis. In: Proceedings IEEE international test conference (ITC), pp 1–10Google Scholar
  7. 7.
    Zhang Z, Chakrabarty K, Wang Z, Wang Z, Gu X (2011) Smart diagnosis: efficient board-level diagnosis and repair using artificial neural networks. In: Proceedings IEEE international test conference (ITC), pp 1–10Google Scholar
  8. 8.
    Amati L (2012) Test and diagnosis strategies for digital devices: methodologies and tools. PhD dissertation, Politecnico di Milano, ItalyGoogle Scholar
  9. 9.
    Pelckmans K et al (2005) Handling missing values in support vector machine classifiers. Neural Netw 18(5):684–692CrossRefzbMATHGoogle Scholar
  10. 10.
    Chechik G et al (2008) Max-margin classification of data with absent features. J Mach Learn Res 9:1–21zbMATHGoogle Scholar
  11. 11.
    Saar-Tsechansky M, Provost F (2007) Handling missing values when applying classification models. J Mach Learn Res 8:1625–1657zbMATHGoogle Scholar
  12. 12.
    Hruschka ER et al (2003) Evaluating a nearest-neighbor method to substitute continuous missing values. Advances in artificial intelligence. Springer, Berlin, pp 723–734Google Scholar
  13. 13.
    Rish I (2001) An empirical study of the naive bayes classifier. IBM New York workshop on empirical methods in artificial intelligence, vol 3(22), pp 41–46Google Scholar
  14. 14.
    Quinlan J (1986) Induction of decision trees. Mach Learn 1(1):81–106Google Scholar
  15. 15.
    Ding Y, Peng X, Fu X (2009) The research of artificial neural network on negative correlation learning. Advances in computation and intelligence. Springer, New York, pp 392–399Google Scholar
  16. 16.
    Prieto B, de Lope J, Maravall D (2005) Reconfigurable hardware implementation of neural networks for humanoid locomotion. Artificial intelligence and knowledge engineering applications: a bioinspired approach. Springer, New York, pp 395–404Google Scholar
  17. 17.
    Vapnik V (1995) The nature of statistical learning theory. Springer, New YorkCrossRefzbMATHGoogle Scholar
  18. 18.
    Chang C-C, Lin C-J (2011) LIBSVM: a library for support vector machines. ACM transactions on intelligent systems and technology, vol 2, pp 27:1–27:27. Software available at

Copyright information

© Springer International Publishing Switzerland 2017

Authors and Affiliations

  1. 1.Huawei TechnologiesSanta ClaraUSA
  2. 2.Department of Electrical and Computer EngineeringDuke UniversityDurhamUSA

Personalised recommendations