Classification of Spam Emails Using Random Forest Algorithm In Comparison With Naive Bayes Algorithm
Keywords:
Machine Learning, Novel Tree Specific Method, Random Forest, Naive Bayes, Spam Filtering, Black List, White List.Abstract
Aim: The main aim of the research is to classify the spam emails using Random Forest over Naive Bayes Algorithm. Materials and Methods: Random Forest algorithm and Naive Bayes algorithm are implemented in this research work. Sample size of n = 20 calculated using G power software, G power value is between 0.59 and 0.9 and determined as 10 per group with pretest power 80%, threshold 0.05% and CI 95%. Result: Random Forest algorithm provides a higher accuracy of 98.33% compared to Naive Bayes algorithm with 88.22% to classify . There is a significant difference between two groups with a significance value of 0.049 (p<0.05). Conclusion: These results show that the performance of the Random Forest algorithm(98.33%) is better than that of the Naive Bayes algorithm(88.22%) in terms of accuracy.