A computational perspective on influence between legislative and media agendas – Medialex

The MEDIALEX project mobilizes methods from computational social sciences, taking advantage of new analytical frameworks that reconcile social science approaches with new computational tools (Cointet and Parasie, 2018; Boelaert, Ollion, 2018; Evans, 2020). The project's methodological challenge is to develop methods derived from automatic language processing, so as to enable finer-grained, large-scale identification of the themes running through public debate. These methods concern, on the one hand, the identification of events on Twitter, and, on the other, the identification of discourses reported in voluminous and heterogeneous corpora (newspapers, television, radio, parliamentary questions and debates, Twitter and Facebook). On the one hand, the method for identifying events on Twitter has been developed and tested by the project partners (Mazoyer et al., 2020). This consists in detecting events based on the thematic similarity of tweets and their temporal proximity. This method identifies themes on a much finer scale than topic modeling methods. On the other hand, the aim is to develop a method for identifying reported discourse in large, heterogeneous textual corpora. The approach will be based on machine learning techniques and will aim to recognize triplets < source, support, propos_rapporté>. The analysis will be based on the use of recent language models (Bert for French, Camembert or Flaubert), which have shown good performance for a similar task with English (Luo et al., 2020).
The MEDIALEX project intends to mobilize these computational methods both to identify global dynamics of influence between parliamentary, media and public actors, and to analyze the dynamics of influence that are established on the occasion of specific bills. During the first year of the project, we will be identifying a number of bills of particular interest in terms of the mechanisms of influence between parliamentary, media and public players. By way of example, a major economic bill and a security-related bill would be relevant to the project.
Finally, the project attaches great importance to integrating a qualitative dimension into quantitative and computational analyses. It is particularly important to study qualitatively how French parliamentarians mobilize social networks and more traditional media. This will inform the more computational analyses, and enable us to develop a set of hypotheses through contact with parliamentary players in particular.

MEDIALEX's vocation is essentially scientific, but given its object and the techniques implemented, we wish to project the results of MEDIALEX into public debate. Indeed, we believe that the project's approach, in the form of a research prototype, will enable us to devise new ways of representing public space and political activity. In this way, MEDIALEX's approach embraces some of our society's major preoccupations concerning the role of Parliament, the production of public policy, the role of the media and the renewal of forms of democracy.
An “open” project. MEDIALEX works in the spirit of open science. The project will not register any patents, the software developed will be under a free open source license, and the data gathered in the project will be made available in open data formats (respecting any constraints of confidentiality, privacy or intellectual property). In addition, MEDIALEX intends to share its work (through datasprints - WP1.3) and results with a wide range of players in the scientific community (in IT and the social sciences) and with players in the ecosystem (journalists, politicians, civil servants, civic tech activists and citizens) with whom the project members already have a significant number of relationships and involvements.
Scientific communication. The scientific dissemination of the project will take the form of A-ranked academic journal articles, the organization of a final conference and the publication of an academic book. The conferences and journals targeted for these scientific publications are Revue française de science politique, Gouvernement et action publique and Parliamentary affairs, for political science; ICWSM (International Conference on Webblogging and Social Media) and IC2S2 (International Conference on Computational Social Science) for computational social sciences; New Media & Society, Big data & Society for sociology and information and communication sciences.
Institutional reflexivity. Finally, MEDIALEX intends to contribute to the discussion on institutional reform by specifically communicating the results to political players, specialized journalists and committed civic tech players. Several participants have important links with parliamentarians and the civic tech world (Olivier Rozenberg, Etienne Ollion, Benjamin Ooghe-Tabanou), which could be put to good use during the planned datasprints and workshops (WP1.3 and WP1.4). On certain questions relating to the mediatization of parliamentary work, project members will be called upon to make public policy proposals.

In a context of destabilization of the public space linked to digital technology, the MEDIALEX project aims to renew the understanding of the dynamics of influence between parliamentary, media and public agendas. Its main objective is to better understand how parliamentarians, the media and the public influence each other in the definition of priority topics for public debate. To achieve this, the project intends to develop new computational methods to track statements in different layers of the public space. By bringing together sociologists, political scientists, computer scientists and computational linguistics researchers, this interdisciplinary project aims to (1) understand the dynamics of influence between the agendas of parliamentarians, the media and the public, (2) develop original methods to identify media events and reported utterances in large heterogeneous corpora, and (3) study the effects of the digitization of the public space on the legislator's ability to impose the topics of public discussion.

The project mobilizes methods from the computational social sciences, taking advantage of new analytical frameworks that reconcile social science approaches with new computational tools. The methodological challenge of the project consists in developing methods from automatic language processing, in order to allow the identification of themes that cross the public debate in a more refined way and on a larger scale. These methods concern the identification of events on Twitter, on the one hand, and the identification of discourses reported in voluminous and heterogeneous corpora (newspapers, television, radio, parliamentary questions and debates, Twitter and Facebook).

The scientific program of MEDIALEX is divided into four work packages. The first one gathers the tasks of coordination, corpus management and dissemination (WP1). The three other work packages explore the influence of parliamentary, media and public agendas in three complementary ways. WP2 considers influence in a structural way, aiming at identifying over the long term which large category of actors (parliamentarians, media, public) manages to impose priority topics of attention on the others. WP3 considers influence at a finer scale, by studying the mechanisms of circulation of discourses between parliamentary, media and public spaces. Finally, WP4 focuses on the interpretations that are produced by the media and the public from parliamentary work (WP4).

The vocation of MEDIALEX is essentially scientific, but given its object and the techniques implemented, the project aims to contribute to public debate. Through "datasprints" and "workshops", the project intends to involve actors and experts around new ways of representing the public space and political activity. MEDIALEX's approach is thus in line with the major concerns of our societies regarding the role of Parliament, the production of public policies, the role of the media and the renewal of the forms of democracy.