Full Event Interpretation

  • Thomas KeckEmail author
Part of the Springer Theses book series (Springer Theses)


The Full Event Interpretation (FEI) is a tagging algorithm based on machine learning. It exploits the unique experimental setup of \(\text {B}\) factory experiments such as the Belle and Belle II experiment. Both experiments operate on the \(\Upsilon (4\text {S})\) resonance, which decays at least \(96 \%\) of the time into exactly two \(\text {B}\) mesons. Conceptually, the event is divided into two sides: The signal-side containing the tracks and clusters compatible with the assumed signal \(\text {B}_{\mathrm {sig}} \) decay the physicist is interested in, e.g. \(\text {B}^{+}\rightarrow \tau ^{+}\nu _{\tau }\); and the tag-side containing the remaining tracks and clusters compatible with an arbitrary \(\text {B}_{\mathrm {tag}} \) meson decay. Figure 4.1 depicts this situation.


  1. 1.
    A.J. Bevan et al., The physics of the B factories. Eur. Phys. J. C 74, 3026 (2014). Scholar
  2. 2.
    T. Keck, The full event interpretation for Belle II. M.A. thesis. KIT, 2014,
  3. 3.
    M. Feindt et al., A hierarchical neurobayes-based algorithm for full reconstruction of B mesons at B factories. Nucl. Instrum. Methods A654, 432–440 (2011). Scholar
  4. 4.
    W.D. Hulsbergen, Decay chain fitting with a Kalman filter. Nucl. Instrum. Methods A552, 566–575 (2005). Scholar
  5. 5.
    W. Waltenberger, W. Mitaroff, F. Moser, RAVE—a Detector-independent vertex reconstruction toolkit. Nucl. Instrum. Methods Phys. Res. Sect. A: Accel. Spectrom. Detect. Assoc. Equip. 581(1–2), 549–552 (2007), Scholar
  6. 6.
    R. Fruhwirth, Application of Kalman filtering to track and vertex fitting. Nucl. Instrum. Methods (1987). Scholar
  7. 7.
    G. M. Amdahl, Validity of the single processor approach to achieving large scale computing capabilities, in Proceedings of the Spring Joint Computer Conference, pp. 483–485, 18–20 April 1967,
  8. 8.
  9. 9.
    K. Kirchgessner, Semileptonic tag side reconstruction. M.A. thesis. KIT, 2012,
  10. 10.
    I. Adachi et al., Belle design study report (2008), arXiv: 0810.4084 [hep-ex]
  11. 11.
    B. Kronenbitter et al., Measurement of the branching fraction of \(B^{+}\rightarrow \tau ^{+}\nu _\tau \) decays with the semileptonic tagging method. Phys. Rev. D 92(5), 051102 (2015). Scholar
  12. 12.
    J. Schwab, Calibration of the full event interpretation for the Belle and the Belle II experiment. M.A. thesis, KIT, 2017,
  13. 13.
    A. Sibidanov et al., Study of exclusive \(B \rightarrow X_{u}l\nu \) decays and extraction of \(|V_{ub}|\) using full reconstruction tagging at the Belle experiment. Phys. Rev. D (2013). Scholar
  14. 14.
    J. Grygier, Search for \(B\rightarrow h\nu \nu \) decays with semileptonic tagging at Belle (2017), arXiv: 1702.03224 [hep-ex]
  15. 15.
    A. Heller, Search for \(B^{+} \rightarrow \ell ^{+} \nu \gamma \) decays with hadronic tagging using the full Belle data sample. Ph.D. thesis. KIT, 2015,
  16. 16.
    M. Huschle, Measurement of the branching ratio of \(B \rightarrow D^{(*)}\tau \nu _\tau \) relative to \(B \rightarrow D^{(*)}\ell \nu _\ell \) decays with hadronic tagging at Belle. Ph.D. thesis. KIT, 2015,
  17. 17.
    Y. Lecun, Y. Bengio, G. Hinton, Deep learning. Nature 521, 436–444 (2015). Scholar
  18. 18.
    A. Santoro et al., A simple neural network module for relational reasoning (2017), arXiv: 1706.01427 [cs.CL]
  19. 19.
    J. Schmidhuber, Deep learning in neural networks: An overview. Neural Netw. (2015). Scholar

Copyright information

© Springer Nature Switzerland AG 2018

Authors and Affiliations

  1. 1.Institute of Experimental Particle PhysicsKarlsruhe Institute of TechnologyKarlsruheGermany

Personalised recommendations