“…There are several studies on Islamophobic and The Use of Twitter's Data, as follows: "Detecting weak and strong Islamophobic hate speech on social media" build multi-class classifiers that distinguish between non-Islamophobic, weak Islamophobic and strong Islamophobic content [10], "Predicting online Islamophobic behavior after #ParisAttacks" in this research collected tweets related to the Paris attacks within 50 hours after the event [11], "Election result prediction using Twitter sentiment analysis" the training dataset, for labeling, using valence aware dictionary and sentiment reasoner (VADER) Next use two algorithms, multinomial Naïve Bayes and SVM [12], "Sentiment analysis about e-commerce from tweet using decision tree , k-nearest neighbor, and Naïve Bayes" use rapidminer to make sentiment analysis by comparing the Decision Tree, K-NN, and Naïve Bayes Classifier and using 10-Fold Cross validation to evaluate the performance of the machine learning model and get the highest results from Naïve Bayes [13], "Analisis Sentimen Pada Review Konsumen Menggunakan Metode Naïve Bayes Dengan Seleksi Fitur Chi Square Untuk Rekomendasi Lokasi Makanan Tradisional" using feature selection the chi-square and classification process using the Naïve Bayes method [14], "An evaluation of SVM and Naïve Bayes with SMOTE on sentiment analysis data set" determine the factors involving the classification of SVM and Naïve Bayes in sentiment classification of problems using SMOTE in the dataset, also comparing the use of 10-fold cross validation with 70:30 split in the test [15].…”