Detection of Bots and Cyborgs in Twitter: A Study on the Chilean Presidential Election in 2017.

The increase of content in social networks, and specially its use in the political environment, has led to the creation and proliferation of autonomous entities commonly known as bots. Bots are programms that performs an automated task over the internet. In this study these entities were initially detected based on a manual analysis carried out on the activity produced by electoral candidates in Chile during the presidential election in year 2017. As a result of this, the need to identify these accounts in an automatic way arose, in order to asses the impact of these accounts in the social network activity during the presidential election in 2017. Various features were extracted in order to train Machine Learning algorithms for the automatic classification task using a set of publicly available data, and other semi-automatic approaches. The models obtain over \(80\%\) in the training stage, but less than \(60\%\) in the testing stage, thus encouraging us to continue to work in other types of representations and models in order to improve the results.