Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2504.18279 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910918911197184 |
|---|---|
| author | Mallick, Neelkamal Hassan, Hadi Kim, D. J. |
| author_facet | Mallick, Neelkamal Hassan, Hadi Kim, D. J. |
| contents | Studies of heavy-quark (charm and beauty) production in hadronic and nuclear collisions provide excellent testing grounds for the theory of strong interaction, quantum chromodynamics. Heavy-quarks are produced predominantly in the initial hard partonic interactions, allowing them to witness the entire evolution process. The charm hadrons are produced in two ways. Firstly, the prompt charm hadrons which are formed from the charm quark hadronization which are produced directly from the initial hard-scatterings or the decay of other excited charm states. Secondly, the nonprompt charm hadrons which are produced from the decay of beauty hadrons. The produced charm hadrons then usually decay to light-flavor hadrons or leptons via two or three body decay. The reconstruction of charm hadrons is challenging due to the large combinatorial background as well as the difficulty of distinguishing between prompt and non-prompt charm hadrons. In this work, we use machine learning models--XGboost and Deep Neural Network--to reconstruct $Λ_c^{+} (udc)$ hadrons via its three body final state decay channel, $Λ_c^{+} \rightarrow pK^0_s$ and $K^0_s \rightarrow π^{+}π^{-}$. Using several experimentally available features of the decay daughters, these models can separate signal from background and identify prompt and nonprompt candidates with nearly 99\% accuracy. This method performs an unbinned track-level reconstruction since the $Λ_c$ candidates are tagged directly from their decay daughters. The necessary data for this study are simulated in pp collisions at $\sqrt{s}=13.6$~TeV using PYTHIA8 (Monash) model. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_18279 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Charm-hadron reconstruction through three body decay in hadronic collisions using Machine Learning Mallick, Neelkamal Hassan, Hadi Kim, D. J. High Energy Physics - Phenomenology High Energy Physics - Experiment Nuclear Theory Studies of heavy-quark (charm and beauty) production in hadronic and nuclear collisions provide excellent testing grounds for the theory of strong interaction, quantum chromodynamics. Heavy-quarks are produced predominantly in the initial hard partonic interactions, allowing them to witness the entire evolution process. The charm hadrons are produced in two ways. Firstly, the prompt charm hadrons which are formed from the charm quark hadronization which are produced directly from the initial hard-scatterings or the decay of other excited charm states. Secondly, the nonprompt charm hadrons which are produced from the decay of beauty hadrons. The produced charm hadrons then usually decay to light-flavor hadrons or leptons via two or three body decay. The reconstruction of charm hadrons is challenging due to the large combinatorial background as well as the difficulty of distinguishing between prompt and non-prompt charm hadrons. In this work, we use machine learning models--XGboost and Deep Neural Network--to reconstruct $Λ_c^{+} (udc)$ hadrons via its three body final state decay channel, $Λ_c^{+} \rightarrow pK^0_s$ and $K^0_s \rightarrow π^{+}π^{-}$. Using several experimentally available features of the decay daughters, these models can separate signal from background and identify prompt and nonprompt candidates with nearly 99\% accuracy. This method performs an unbinned track-level reconstruction since the $Λ_c$ candidates are tagged directly from their decay daughters. The necessary data for this study are simulated in pp collisions at $\sqrt{s}=13.6$~TeV using PYTHIA8 (Monash) model. |
| title | Charm-hadron reconstruction through three body decay in hadronic collisions using Machine Learning |
| topic | High Energy Physics - Phenomenology High Energy Physics - Experiment Nuclear Theory |
| url | https://arxiv.org/abs/2504.18279 |