Using natural language technology to measure Mass media reactions in the election context Daniela Gîfu 1 , Dan Cristea 1,2 , 1 „Alexandru Ioan Cuza“ University, Faculty of Computer Science, 16, General Berthelot St., 700483, Ia i ş {daniela.gifu, dcristea}@info.uaic.ro 2 Institute for Theoretical Computer Science, Romanian Academy - Ia i branch, 2, T. ş Codrescu St., 700481, Ia i ş Abstract. This paper presents a computational tool, PEDANT, based on natural language processing (NLP) techniques for the interpretation of the political discourse in the print media. This application considers the 2009 presidential campaign in Romania. The concept behind this method is that the manner in which individuals speak and write, with the aim to deliver a certain image to the public, is an opened window towards their emotional and cognitive worlds. In other words, the vocabulary betrays the author’s sensibility. By emphasizing the emotional component at the level of discourse, voters identify with the speaker, who becomes the personification of their common ideals. Our investigation is intended to give support to researchers, specialists in political sciences, to journalists and election’s staff, being helpful mainly in their exploration of the political campaigns, in their intend to measure reactions with respect to the developments in the political scene. Keywords: political discourse, natural language processing, elections, newspaper, semantic analysis, journalist. 1 Introduction It is known that the text of the press may be of inconvenience or can be useful to those that become the subject of the press. Most of the time the subjects are the ones in political power of that time. [6]. The motivation for our study relies on the need for objectivity in the interpretation of the political language, situated at the intersection of three important symbolic spaces: the political space, the public space and the communicational space, as well as on the need to measure to what extend a discourse can influence its direct receptor, the electorate and in what ways. [5]. Among many attributes the political discourse has, we were interested in the lexicon and its interpretation in a range of semantic coordinates. The final objective of our research is to develop a computational framework able to offer to researchers in Mass media, political sciences, to political analysts, to the public at large (interested to consolidate their options before elections), and, why not, even to politicians