A New Framework For Privacy Setting Using Machine Learning Algorithms Using Bag Of Words Classifier

Abstract

The primary goal of this research project is to identify and filter spam terms used in communication between two or more people. The dataset for training and testing the proposed prediction models was built using short messaging service (SMS) messages including some spam and ham terms, a minimum of two characteristics, and a sample size of 160 with two groups with a g-power value of 80%. And the banking customer loan data were collected from various web sources with recent study findings and threshold 0.05%, confidence interval 95% mean and standard deviation.  The framework was developed using the bag of words classifier with Multinomial naive bayes (MNB) algorithm and is compared with Support vector machine (SVM) learning algorithm.  The retrieval accuracy rate of MNB algorithm is (97.77%) and the same for SVM learning algorithm is (90.87%) with two tailed significant values p is 0.001 is less than 0.05 significant level with 95% confidence interval.: This research confirms that the effectiveness of the MNB algorithm results is more accurate than the SVM algorithm which was developed using python.