1: \begin{abstract}

2: This work presents a Bayesian algorithm to fit the recombination and

3: collisionally excited line spectra of gas photoionized by clusters

4: of young stars. The current model consists in fourteen dimensions:

5: two electron temperatures, one electron density, the extinction coefficient,

6: the optical depth on the $HeI$ recombination lines and nine ionic

7: species. The results are in very good agreement with those previously

8: published using the traditional methodology. The probabilistic programming

9: library PyMC3 was chosen to explore the parameter space via a NUTs

10: sampler. These machine learning tools provided excellent convergence

11: quality and speed. The primordial helium abundance measured from a

12: multivariable regression using oxygen, nitrogen and sulfur was $Y_{P,\,O-N-S}=0.243\pm0.005$

13: in agreement with a standard Big Bang scenario.

14: \end{abstract}

15: