In recent times, political behavior, from the act of voting to the participation of citizens in politics, has changed significantly. The proliferation and growth of Information and Communication Technologies (ICTs) has provided new and powerful tools to all stakeholders. In particular, social media allow a two-way communication channel between political parties and the electorate. Under these circumstances, accurate segmentation of electoral markets is essential for the development of campaign messages. To enable personalized one-to-one dialogue, it is necessary to characterize each user. However, this poses two major challenges. On the one hand, the degree of subjectivity in the political domain is difficult to determine because a fact can be considered positive or negative depending on the point of view. On the other hand, political polarization and partisanship, which refers to the fact that citizens are strongly biased in favor of certain political parties while strongly disagreeing with others. The goal of this work is to integrate and validate some of the Natural Language Processing (NLP) technologies developed and tested by the participating researchers in previous projects for the deployment and optimization of a commercial software platform for political microtargeting through author and user profiling.