Inferring the phylogeny of Languages

In the last decade, bayesian statistics have proven their efficiency in the inference of phylogenetic trees for families of languages. Despite the recent improvements of the models, the studies are still based on cognacy informations, the constitution of which is a long, difficult and debatable work; hence inducing a bias.
We will present a model specific to the study of sign language that will allow us, not only to remove this bias, but also to model phonological evolution. This model requires the use of several advanced numerical tools that we will present.