Seleção e agrupamento de características para classificação de agentes automatizados em redes sociais

Abstract

The increasing use of social networks has also led to the rise of the use of automated means to generate or propagate content, the dissemination of such content can be harmful to the social sphere. With the discovery of a new range of problems for the identification of such automated media, this paper aims to analyze new approaches to information handling and new machine learning models for social networking profiles providing automated identification of the use of such practices. First, an analysis of the database made available in the profiles is performed, with this it is possible to analyze which characteristics are relevant to the classification of the profiles. Using the analysis of grouping techniques are applied to analyze the results of the models using this approach to the problem of profile classification. Soon after, models are trained using the approaches proposed in this work and using the data from a database, where the data were processed for better use in the models at the end of the training, the results will be compared with the state of the art.