Naive Bayesian Algorithm for Spam Classification Based on Random Forest Method
Lu Wang, Guohui Zeng, Bo Huang · Journal of Physics Conference Series · 2020
Abstract Aiming at the problem of dimensionality disasters in the spam filtering system, a spam filtering model based on content feature selection random forest algorithm is proposed. The method combines the Naïve Bayesian algorithm and effectively reduces the influence of redundant information on classification performance through feature selection filtering of Gini impurity. The improved algorithm takes advantage of decision tree integration and naive Bayesian algorithm. The experimental results show that compared with the naive Bayesian algorithm, the improved algorithm has been improved in the classification performance indicators of spam filtering, which is more feasible and effective in practical applications.