Terrorism leads to a huge social, economic, and psychological crisis due to the various terrorist activities that affect various countries, causing loss of life, economy, and property. These attacks include drilling sessions, training, preparation, execution, threats among the common people, and collaboration between these terrorists. Terrorist attacks are unpredictable, and we don’t know what type of attack will be done. To counter these frequent terrorist attacks, the implementation of ensemble learning for measuring the intensity of terrorist attacks is done. In our research, we propose a classification framework based on ensemble learning for classifying and predicting terrorist organizations. Based on analysis of terrorist organization activities in GTD from 2000 to 2022 and using the ensemble learning approach, we construct classification and prediction models of terrorist attacks, namely, decision tree, random forest, Catboost, and XGBoost, and utilized a TF-IDFVectorizer method to evaluate the performance and stability of the proposed model. In this research various methods such as bagging, boosting, stacking, etc., are applied on Global Terrorism Database (GTD). The XGBoost and CatBoost models achieved the best accuracies (92.16% and 90.82%, respectively) with the highest attack frequencies.