Comparison of accuracy in naive bayes algorithm with gradient boosting algorithm in detection of fake news.

Autor: Reddy, N. Dinesh Kumar, Pramila
Předmět:
Zdroj: AIP Conference Proceedings; 2023, Vol. 2822 Issue 1, p1-8, 8p
Abstrakt: To perform accurate Fake News Detection using Gradient Boosting and compare its textual property accuracy with Naive Bayes algorithm. The analysis for fake news detection in this proposed research was done using machine learning algorithms such as the Gradient Boosting algorithm (N=311) and Naive Bayes algorithm (N=311) with G power 80 % and alpha value 0.05. The accuracy of the Gradient Boosting algorithm appears to be 99.54 %, and the accuracy of the Naive Bayes algorithm appears to be 95.23%. With a significance value of p= 0.000 (2-tailed) for accuracy and 0.125 for precision, there is statistically no significant difference between the sample groups. The Gradient Boosting algorithm appears to be more accurate than the Naive Bayes algorithm in identifying fake news. [ABSTRACT FROM AUTHOR]
Databáze: Complementary Index