Study of D-meson decay-topology selection with machine learning in pp collisions at $\sqrt{s} = 13$ TeV for ALICE in Run 2

The ratio $D_s^+ / D^+$ is measured as a function of transverse momentum ($p_T$) in the range $2 < p_T < 12$ GeV/c in high-multiplicity pp collisions at a centre-of-mass energy $\sqrt{s} = 13$ TeV collected with the ALICE detector at the LHC. The D mesons were reconstructed from the decay channel $D_s^{+}(D^+) \rightarrow \phi \pi^+ \rightarrow K^{-} K^{+} \pi^+$. A machine learning multi-class classification algorithm was employed to make preselections at the point where the efficiencies for the prompt $D_s^+$ and $D^+$ mesons were equal, enabling the ratio $D_s^+ / D^+$ to be immediately acquired without first having to compute individual cross-sections and thus significantly reducing its associated systematic uncertainty. As a proof of method, this analysis gives results compatible with previous measurements at the same centre-of-mass energy and with a centre-of-mass energy at $\sqrt{s} = 5.02$ TeV.