A machine learning-based methodology for pulse classification in dual-phase xenon time projection chambers
Abstract Machine learning techniques are now well established in experimental particle physics, allowing detector data to be analyzed in new and unique ways. The identification of signals in particle observatories is an essential data processing task that can potentially be improved using such metho...
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
SpringerOpen
2022-06-01
|
| Series: | European Physical Journal C: Particles and Fields |
| Online Access: | https://doi.org/10.1140/epjc/s10052-022-10502-x |
| _version_ | 1811230753101447168 |
|---|---|
| author | P. Brás F. Neves A. Lindote A. Cottle R. Cabrita E. Lopez Asamar G. Pereira C. Silva V. Solovov M. I. Lopes |
| author_facet | P. Brás F. Neves A. Lindote A. Cottle R. Cabrita E. Lopez Asamar G. Pereira C. Silva V. Solovov M. I. Lopes |
| author_sort | P. Brás |
| collection | DOAJ |
| description | Abstract Machine learning techniques are now well established in experimental particle physics, allowing detector data to be analyzed in new and unique ways. The identification of signals in particle observatories is an essential data processing task that can potentially be improved using such methods. This paper aims at exploring the benefits that a dedicated machine learning approach might provide to the classification of signals in dual-phase noble gas time projection chambers. A full methodology is presented, from exploratory data analysis using Gaussian mixture models and feature importance ranking to the construction of dedicated predictive models based on standard implementations of neural networks and random forests, validated using unlabeled simulated data from the LZ experiment as a proxy to real data. The global classification accuracy of the predictive models developed in this work is estimated to be >99.0%, which is an improvement over conventional algorithms tested with similar data. The results from the clustering analysis were also used to identify anomalies in the data caused by miscalculated signal properties, showing that this methodology can also be used for data monitoring. |
| first_indexed | 2024-04-12T10:34:21Z |
| format | Article |
| id | doaj.art-0589d7c608f74fe2860744a58b5fe76f |
| institution | Directory Open Access Journal |
| issn | 1434-6052 |
| language | English |
| last_indexed | 2024-04-12T10:34:21Z |
| publishDate | 2022-06-01 |
| publisher | SpringerOpen |
| record_format | Article |
| series | European Physical Journal C: Particles and Fields |
| spelling | doaj.art-0589d7c608f74fe2860744a58b5fe76f2022-12-22T03:36:46ZengSpringerOpenEuropean Physical Journal C: Particles and Fields1434-60522022-06-0182611710.1140/epjc/s10052-022-10502-xA machine learning-based methodology for pulse classification in dual-phase xenon time projection chambersP. Brás0F. Neves1A. Lindote2A. Cottle3R. Cabrita4E. Lopez Asamar5G. Pereira6C. Silva7V. Solovov8M. I. Lopes9Laboratório de Instrumentação e Física Experimental de Partículas (LIP), University of CoimbraLaboratório de Instrumentação e Física Experimental de Partículas (LIP), University of CoimbraLaboratório de Instrumentação e Física Experimental de Partículas (LIP), University of CoimbraDepartment of Physics, University of OxfordLaboratório de Instrumentação e Física Experimental de Partículas (LIP), University of CoimbraLaboratório de Instrumentação e Física Experimental de Partículas (LIP), University of CoimbraLaboratório de Instrumentação e Física Experimental de Partículas (LIP), University of CoimbraLaboratório de Instrumentação e Física Experimental de Partículas (LIP), University of CoimbraLaboratório de Instrumentação e Física Experimental de Partículas (LIP), University of CoimbraLaboratório de Instrumentação e Física Experimental de Partículas (LIP), University of CoimbraAbstract Machine learning techniques are now well established in experimental particle physics, allowing detector data to be analyzed in new and unique ways. The identification of signals in particle observatories is an essential data processing task that can potentially be improved using such methods. This paper aims at exploring the benefits that a dedicated machine learning approach might provide to the classification of signals in dual-phase noble gas time projection chambers. A full methodology is presented, from exploratory data analysis using Gaussian mixture models and feature importance ranking to the construction of dedicated predictive models based on standard implementations of neural networks and random forests, validated using unlabeled simulated data from the LZ experiment as a proxy to real data. The global classification accuracy of the predictive models developed in this work is estimated to be >99.0%, which is an improvement over conventional algorithms tested with similar data. The results from the clustering analysis were also used to identify anomalies in the data caused by miscalculated signal properties, showing that this methodology can also be used for data monitoring.https://doi.org/10.1140/epjc/s10052-022-10502-x |
| spellingShingle | P. Brás F. Neves A. Lindote A. Cottle R. Cabrita E. Lopez Asamar G. Pereira C. Silva V. Solovov M. I. Lopes A machine learning-based methodology for pulse classification in dual-phase xenon time projection chambers European Physical Journal C: Particles and Fields |
| title | A machine learning-based methodology for pulse classification in dual-phase xenon time projection chambers |
| title_full | A machine learning-based methodology for pulse classification in dual-phase xenon time projection chambers |
| title_fullStr | A machine learning-based methodology for pulse classification in dual-phase xenon time projection chambers |
| title_full_unstemmed | A machine learning-based methodology for pulse classification in dual-phase xenon time projection chambers |
| title_short | A machine learning-based methodology for pulse classification in dual-phase xenon time projection chambers |
| title_sort | machine learning based methodology for pulse classification in dual phase xenon time projection chambers |
| url | https://doi.org/10.1140/epjc/s10052-022-10502-x |
| work_keys_str_mv | AT pbras amachinelearningbasedmethodologyforpulseclassificationindualphasexenontimeprojectionchambers AT fneves amachinelearningbasedmethodologyforpulseclassificationindualphasexenontimeprojectionchambers AT alindote amachinelearningbasedmethodologyforpulseclassificationindualphasexenontimeprojectionchambers AT acottle amachinelearningbasedmethodologyforpulseclassificationindualphasexenontimeprojectionchambers AT rcabrita amachinelearningbasedmethodologyforpulseclassificationindualphasexenontimeprojectionchambers AT elopezasamar amachinelearningbasedmethodologyforpulseclassificationindualphasexenontimeprojectionchambers AT gpereira amachinelearningbasedmethodologyforpulseclassificationindualphasexenontimeprojectionchambers AT csilva amachinelearningbasedmethodologyforpulseclassificationindualphasexenontimeprojectionchambers AT vsolovov amachinelearningbasedmethodologyforpulseclassificationindualphasexenontimeprojectionchambers AT milopes amachinelearningbasedmethodologyforpulseclassificationindualphasexenontimeprojectionchambers AT pbras machinelearningbasedmethodologyforpulseclassificationindualphasexenontimeprojectionchambers AT fneves machinelearningbasedmethodologyforpulseclassificationindualphasexenontimeprojectionchambers AT alindote machinelearningbasedmethodologyforpulseclassificationindualphasexenontimeprojectionchambers AT acottle machinelearningbasedmethodologyforpulseclassificationindualphasexenontimeprojectionchambers AT rcabrita machinelearningbasedmethodologyforpulseclassificationindualphasexenontimeprojectionchambers AT elopezasamar machinelearningbasedmethodologyforpulseclassificationindualphasexenontimeprojectionchambers AT gpereira machinelearningbasedmethodologyforpulseclassificationindualphasexenontimeprojectionchambers AT csilva machinelearningbasedmethodologyforpulseclassificationindualphasexenontimeprojectionchambers AT vsolovov machinelearningbasedmethodologyforpulseclassificationindualphasexenontimeprojectionchambers AT milopes machinelearningbasedmethodologyforpulseclassificationindualphasexenontimeprojectionchambers |