Big Data Analysis Techniques Applied to the NA62 Experiment at CERN
Lead Research Organisation:
Lancaster University
Department Name: Physics
Abstract
Abstracts are not currently available in GtR for all funded research. This is normally because the abstract was not required at the time of proposal submission, but may be because it included sensitive information such as personal details.
People |
ORCID iD |
| Giuseppe Ruggiero (Primary Supervisor) | |
| Joseph Carmignani (Student) |
Publications
Cortina Gil E
(2019)
Searches for lepton number violating K+ decays
in Physics Letters B
Studentship Projects
| Project Reference | Relationship | Related To | Start | End | Student Name |
|---|---|---|---|---|---|
| ST/P006795/1 | 30/09/2017 | 29/09/2024 | |||
| 2039270 | Studentship | ST/P006795/1 | 01/12/2017 | 30/05/2022 | Joseph Carmignani |
| Description | I've implemented a machine learning algorithm based on a Neural Net (NN) that learn from data by doing a simple mapping between input and output to classify by predictions. The objectives achieved are the following: - NN algorithm works well enough, can learn the physics features and shows a higher performance even from Raw Data (Low Level Variables). -NNs are flexible we can adjust the architecture to adapt with our data. - Features Engineering is a must if we want to take full advantage of NN potentials (in Parameter Space). - The predictions of NN showed a stability over the runs and outperformed a classic discriminant based on logarithmic likelihood. |
| Exploitation Route | Future Work will use the NN for Background rejection in addition to: -We must work on a MC sample to compare with data as well. -Development of a more general algorithm to handle fake tracks built from pileup and Kaon hits in the NA62 experiment at CERN. - Study of intensity effects over accidental beam activity build up in sub-detectors. - Implementation of the NN and other MVA methods to the main analysis. - Investigation of other areas of the analysis that could benefit of a Multi-Variate-Analysis (i.e. tracking in the STRAW, Calorimetery and definition of the signal region). |
| Sectors | Digital/Communication/Information Technologies (including Software) Education Other |