Customer satisfaction questionnaires are a rich and strong source of information for companies to seek loyalty, customer and client retention, optimize resources, and repurchase products. Several advanced machine learning and statistical models have been employed to estimate the customer satisfaction score; however, there is not a single model that can yield the best result in all situations. Ensembles of regression techniques have demonstrated their effectiveness for various applications, where the success of these models lies in the construction of a set of single models. We perform an experimental study using a real database of 129,890 samples from airline companies, in order verify the benefits of ensemble models for predicting customer satisfaction. Accordingly, the present paper evaluates the BAGGING ensemble model using the well-renowned k-nn algorithm as the base learner. The obtained results indicate that the BAGGING ensemble performs better than the single classifier in terms of RMSE and MAE.Bayes, decision tree, logistic regression, and support vector machine) on a dataset acquired from a contact center. The prediction task was addressed as a classification problem, where the best model obtained an accuracy of 66%. Roy et al.[4] employed naïve Bayes, multiclass classifier, k-star, and IBK (Instance-Based learning with parameter K) as classifiers models for predicting customer satisfaction from a database constructed from a customer survey conducted by the San Francisco International Airport. Aktepe et al. [5] showed that using classifier algorithms combined with programming software and structural equation modeling is able to analyze the level of customer satisfaction and loyalty. Farhadloo et al. [6] analyzed the satisfaction of customers using reviews from different states parks in California. These reviews were left by real visitors on TripAdvisor.com. Grigoroudis and Politis [7] suggested that the customer satisfaction problem can be seemed as a multicriteria evaluation problem. Thus, they proposed the MUlticriteria Satisfaction Analysis system (MUSA) that uses ordinal regression techniques. Bockhorst et al.[8] developed a hybrid customer satisfaction system by integrating a linear ranking sub-model and a non-linear isotonic regression. The system was trained on a database constructed by using phone calls from five surveys. Experimental results revealed that the proposed model is better than standard regression techniques. Other algorithms that have also been employed in the customer satisfaction analysis are the CART (Classification And Regression Tree) algorithm [9], artificial neural network approaches [10,11,12], the principal component analysis [13], and the support vector machine algorithm [14].Although previous studies conclude that data mining and machine learning techniques can be successfully used for prediction of customer satisfaction, there is not an overall best algorithm for dealing with customer satisfaction problems. Consequently, ensembles models have emerged to exploit the differ...