Hate speech encompasses expressions that attack, intimidate, humiliate, discredit, or incite violence based on factors such as race, religion, ethnicity, nationality, age, sexual orientation, social class, or disability, posing a significant threat to freedom of expression and tolerance. This type of content primarily spreads through online and mainstream media channels.
The use of electronic communication tools, particularly the Internet and social media platforms, has significantly amplified the visibility and spread of hate speech. Therefore, monitoring and addressing this issue in online environments has become imperative.
The primary objective is to identify and mitigate hate speech while enhancing data collection mechanisms to support efficient police resource allocation. PRESERVE will leverage big data analytics and AI to monitor and assess the risk of hate speech targeting specific communities.
The project will implement advanced AI-driven technologies to monitor and evaluate the reach of specific actors. It will also correlate and analyse multimedia data from social media, news reports, past incidents, and geospatial information.
The PRESERVE toolkit will provide LEAs with a powerful solution to detect, analyse, and respond to hate speech incidents effectively, ensuring better protection for at-risk communities and individuals.
Advanced AI-driven technologies employed include NLP (Natural Language Processing) & NLU (Natural Language Understanding), UEBA, Computer Vision, and Graph Models.