This paper presents the identification of Spam and ham messages using supervised machine learning algorithms Random forest Classifier, and Logistic Regression algorithms and Analyzes how each filter performs when detecting Ham and Spam. A spam message is a big issue in mobile communication to reduce this effective spam detection techniques should be built Preprocessing is done using the NLTK library with various Stemming Algorithms, Word clouds are used and tokenizing is also performed. The data set is divided into two categories for training and testing the classifiers . the results demonstrated that the performance of Random Forest is better than Logistic Regression. Random forest achieved a better accuracy of 97%.