Uniform and normal laws of discrete strings
Sébastien Rebecchi · 2008
We propose in this paper the generalization of concept of statistical distribution to the space of discrete strings, this more particularly for uniform and normal laws. These definitions allow statistical handling of sets of discrete strings, via generative and predictive algorithms. We also propose a method of estimation of parameters of distributions of discrete strings, applied in this paper to the estimation of standard deviation of the normal law, and giving encouraging results. This work is a first trend toward a general statistical framework for recognition of structured patterns.