View web version

Techniques and Services for Identifying Radicalism

and Extremism in Social Media

Which software services could be developed and used for radical and extremist content detection in social media?

Reaching a deeper understanding of the effectiveness of communication strategies is the

first step to designing a counter and alternative narratives campaign aimed at countering

radicalization and extremism. 

 

Based on the expertise of the H2020 Trivalent project, the PARTICIPATION project has developed a multi-dimensional complex model with the aim of understanding radicalization and identifying new trends. The project has analyzed four typologies of violent extremism: far-right, far-left, separatism, and religious extremism. 

 

For a more detailed discussion of the methodology and findings, we refer the interested reader to Far-right, far-left, separatism and religious extremism: comparative desk research on drivers (Marinone et al., 2021) for a more detailed discussion of the methodology and findings. The paper provides an overview of the drivers of radicalization and violent extremism that were identified.

Intelligent engine for analysis of online and offline content

Over the past few months, our research has focused on the methods for detecting and monitoring radicalism and extremism in social media, in order to evaluate the effectiveness of counter-narrative campaigns, as well as developing ICT tools and services for these purposes.

 

One of the objectives of PARTICIPATION is to develop information and communication technology tools for providing effective and automated ways to monitor and evaluate counter-narrative campaigns. Thus, the developed software is operated through an intelligent engine that captures, analyzes, enriches, stores and contextualizes social network interactions that express radical or extremist content.

 

To this end, previous works which address the automatic identification of extremism drivers have been studied, putting the focus on how these can be used to monitor radicalization and other campaigns. In light of this analysis, Techniques and services for identifying radicalism and extremism in social media report describes how the insights gained can be transformed into the definition of several analysis processes that enrich and contextualize the captured data. The development of the intelligent engine follows open linked data conventions to model and process all its internal representations. In doing so, this document illustrates the definition of several vocabularies and taxonomies that are used to model the data. The use of this type of representations offers a common and accessible framework to capture, share, and process the data produced by the intelligent engine.

Linguistic processing and Semantic modelling    

In regard to the analyses performed by the intelligent engine, both the Linguistic Inquiry and Word Count (LIWC) (Pennebaker, 2011) and the Moral Foundations Theory (MFT) (Graham et al., 2009) are exploited, offering additional knowledge with which the captured data is enriched. After adding these psychological drivers and measures, the captured messages can be further analyzed, considering vital contextual information such as elicited emotions, personal drivers, and moral values from the text.

 

In order to fully encompass the mentioned analyses in a way that can be useful for external

users, a pipeline-based architecture has been defined that separates the different steps

through a multi-step process.This system is developed following the semantic web framework, with a focus on usability, extensibility, and compatibility. The developed

system stores all captured and analyzed data in two different databases.  In addition, a new tool is provided which offers a wide range of visualizations and summary modules of the data. This web tool allows users to navigate the dataset in a simple manner, which is useful for monitoring trends and different phenomena in radical and extremist contexts.

The Techniques and services for identifying radicalism and extremism in social media report is now available for download on the project’s website. The results of this study will be used to strengthen the analysis of on-line extremist recruitment and propaganda in further project analysis.

If you want to know more about the methodology, drop us a line at press@participation-in.eu.

About PARTICIPATION

 

PARTICIPATION is a research and innovation project that unites 15 partners in one effort to analyze and prevent radicalization aimed at increasing awareness of at-risk demographics and encouraging resilience through preventative, countering, and de-radicalization approaches. This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 962547.

 

Date of the project: 01/12/2020 – 01/12/2023

Would you like to know more about PARTICIPATION?

Don’t miss the progress of our journey!

Partners

View web version | Unsubscribe

You have received this e-mail at [Email[ because you are subscribed to CESIE newsletter.
We recently updated our Privacy Policy. Renew your consent to data collection here.
If you do not want to receive our newsletters anymore, please click here to unsubscribe.