Classification of Super-Kamiokande atmospheric neutrino events by using neural network
Abstract We present a new event classification method based on neural network for multi-GeV multi-ring samples from Super-Kamiokande (SK) atmospheric neutrino observation. Identifications of neutrino flavors are more difficult for multi-ring events due to several hadrons generated in deep inelastic...
Saved in:
| Published in | Journal of physics. Conference series Vol. 1468; no. 1; p. 12161 |
|---|---|
| Main Authors | , |
| Format | Journal Article |
| Language | English |
| Published |
01.02.2020
|
| Online Access | Get full text |
Cover
Loading…
| Summary: | Abstract
We present a new event classification method based on neural network for multi-GeV multi-ring samples from Super-Kamiokande (SK) atmospheric neutrino observation. Identifications of neutrino flavors are more difficult for multi-ring events due to several hadrons generated in deep inelastic scattering. We employed neural network to optimize the classification from a combination of observed variables. Compared with the conventional method, significant improvements of the
v
e
selection efficiency is confirmed with the comparable misidentification probabilities. Sensitivity to neutrino mass hierarchy can be improved with more statistics of
v
e
samples. |
|---|---|
| ISSN: | 1742-6588 1742-6596 |
| DOI: | 10.1088/1742-6596/1468/1/012161 |