The general objective of the project is to contribute to the improvement of the capacities of State authorities to identify, analyse, monitor and evaluate hate speech in order to design coordinated actions against hate speech motivated by racism, xenophobia, islamophobia or anti-Semitism.

To this end, the project will be based on a methodology for exchanging experiences, identifying good practices at European level and mutual learning.

1. Identification and development of indicators on hate speech on the web

The aim is to develop an action protocol containing a system of indicators, with search criteria, on speeches that promote racism, xenophobia and hatred on the net. The system will also include early warning indicators to assess the intensity, severity, distribution, and potential impact of hate speech, in order to establish recommendations for action to prevent possible discriminatory incidents or hate crimes. The following activities will be carried out within the framework of this action:

  • Identification of experiences and good practices that have been developed in the European Union and that allow progress in the development of the protocol and the system of indicators. This identification of experiences will serve as a basis for the discussion and classification of early warning indicators.
  • First phase of the development of indicators. Typification of key factors for the analysis of hate speech on the Internet and social networks. The project’s Advisory Group will be in charge of identifying key words, phrases, concepts, etc. as well as specific web pages or profiles of social networks (especially Twitter and Facebook) that are identified as hate speech in four areas: racism, xenophobia, islamophobia and anti-Semitism. These four motives related to hate speech have been chosen for a comparative perspective.
  • Second phase of indicator development. It will be based on the construction of a system of early warning indicators to assess the intensity, severity, distribution, and potential impact of hate speech, in order to establish recommendations for action based on them and to prevent possible discriminatory incidents or hate crimes. This system of indicators (b and c) will serve as the basis for the development of the computer tool.

2. Development of the Computer Tool

The aim of this work package is to program a computer algorithm for the identification, monitoring and analysis of online hate speech (especially on Twitter and Facebook). A functionalities document will be elaborated based on the search criteria and indicators developed in the previous phase (1). For the design and programming of the algorithm, the criteria will be translated into search items and the identification of messages containing possible indicators of hate speech.

3. Design of shared strategies

Synergies between institutional actors and key actors (including communities affected by hate speech and civil society) will be strengthened in order to design shared strategies and combat hate speech in the network.

4. Cooperative learning

The development of workpackages 2, 3 and 4 is based on the methodology of cooperative or mutual learning. This methodology proposes the use of group work to improve collective learning. In this model there is, therefore, a double objective: to learn the task, through the applied methodology and to make sure that all the members of the group do it. To this end, workpackages 2, 3 and 4 provide for the constitution of different working and validation groups that facilitate face-to-face interaction and mutual support. In these interactions the participants consult each other, share resources, exchange materials and information, comment on the work that each person is producing, propose modifications, etc., in a climate of responsibility, interest, respect and trust.