Classification of Fraud Websites using Linear regression Algorithm and Recurrent Neural Network Algorithm with Improved Accuracy

Le résumé fourni par la source

The purpose of this work is to improve the detection of fraud websites using Novel Linear Regression Algorithm and Recurrent Neural Network Algorithm. Materials and Methods: Novel Linear regression Algorithm and Recurrent Neural Network Algorithm is executed with varying training and testing splits for predicting the detection of fraud websites. This research utilizes Novel Linear regression Algorithm and Recurrent Neural Network Algorithm with various grounding categories to predict the detection of extremist reviews. The data set in this article utilizes kaggle website data csv file repositories and number of samples (N=11053) The Gpower test used is about 85% (g power setting parameters: α=0.05 and power=0.85). Results: Novel Linear regression Algorithm (98.15%) has the increased accuracy over Recurrent Neural Network Algorithm (97.22%) with an independent samples-t-test significance value of p=0.01 (p<0.05) and there exists a statistical significant difference between the two algorithms. Conclusion: When compared to Recurrent Neural Network Algorithm accuracy, Novel Linear regression Algorithm has higher accuracy.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.