A Prior Distribution for Anti-spam Statistical Bayesian Model

This paper deals with Bayesian models applied to anti-spam. In most anti-spam related researches, authors assume that the probability of spam message is equal to 0.5, which is unrealistic. This pushes us to define a prior and a posterior probability laws to enhance the spam detection and increase the reliability decision. This work differs from previous results using the Bayesian approach for the anti-spam issue, especially through refinement and enhancement of various probability laws.