Creación de un modelo de minería de datos que identifica sentimientos, de los clientes de la banca privada con calificación de riesgo mayor a AA de la ciudad de Guayaquil con datos basados en Twitter.

Abstract

Within this research project is the creation of a data mining model that could perform a sentiment analysis based on data about Ecuadorian private banking which will have predictive power to assign values of "Positive", "Negative" or "Neutral" to the comments that are processed by it. This will be done by making use of programming tools with which data cleaning, natural language processing, and finally in which machine learning algorithms can be applied to perform the training of the model. This is in order to be able to measure the satisfaction levels of users of Ecuadorian private banks automatically and whose results can then be visualized within a data visualization tool. The objectives of this research project include the creation of the data set to be used, the design of the data mining model and its creation. The project ends with the comparison of the performances of the different machine learning classification algorithms, the selection of the best of these for the development of the model and the presentation of the results of the predictive capability of the developed model.