Probalistic model of spam filter system
Yanzhi Xiong, Tianlan Wei · Applied and Computational Engineering · 2023
Numerous tasks have been carried out using probabilistic reasoning, including picture recognition, computer diagnosis, stock price prediction, movie recommendation, and cyber intrusion detection. But until recently, the breadth of probabilistic programming was constrained (partly because of the low processing power), and the majority of inference methods had to be created manually for every job. Nevertheless, in 2015, 3D representations of human faces were created from 2D photographs of such faces using a 50-line probabilistic computer vision software. The inference approach of the software, which was developed in Julia using the Picture package, was based on inverted graphics. "What used to require thousands of lines of code" might now be accomplished in just 50. Probabilistic computing is a method to create systems that help us make decisions in the face of uncertainty. In this paper we propose a spam-filtering system. The novelty of our spam-filtering system is that we utilize probabilistic programming to improve spam-filtering reasoning systems. We implement several factors from email to establish a model to get the output. Our preliminary results show that we successfully accomplish a spam-filtering system. More problem of classification of spam filter can be described in a probabilistic modelling language in our future work.