Fake News Detection System using Logistic Regression and Compare Textual Property with Support Vector Machine Algorithm

Abstract

To perform accurate Fake News Detection using Logistic Re gression (LR) and compare textual property accuracy wi th Support Vector Machine (SVM) algorithm. Materials and Method: The analysis for fake news detection in this proposed research was done us i ng machine learning algo rithms such a s the LR al go rith m <tex>$\boldsymbol{(\mathrm{N}=311)}$</tex> and SVM algorithm <tex>$\boldsymbol{(\mathrm{N}=311)}$</tex> with G power 80 &#x0025; and alpha value 0.05. Results: The accuracy offake news was analyzed using the LR and SVMalgorithms. The accuracy of the LR algorithm appears to be 95.12 &#x0025;, and the accuracy of the SVM algorithm appears to be 91.68 &#x0025;. With a significance value of 0.079 for accuracy and 0.125 for precision, there is a statistically significant value between th e s am pl e groups. Conclusion: The LR algorithm appears to be more accurate th an the SVM algo rithm in identifying whether the news is fake or not.