A web framework for information aggregation and management of multilingual hate speech.

Area of Science:

• Computational Social Science
• Natural Language Processing
• Web Framework Development

Background:

• Social media generates vast user-generated content, enabling public discourse but also hosting hateful content.
• Manual moderation of large-scale online content is infeasible due to volume and complexity.
• Hate speech online can escalate to real-world actions, necessitating effective detection and management tools.

Purpose of the Study:

• To present and evaluate a web framework for collecting, analyzing, and aggregating multilingual textual content from online sources.
• To provide accessible content analysis tools for journalists, academics, and the public, regardless of technical expertise.
• To facilitate hate speech detection and sentiment analysis in multiple languages.

Main Methods:

• Development of a web framework with backend functionality for content collection, monitoring, and semantic analysis.
• Integration of machine learning models and rule-based algorithms for hate speech and sentiment analysis.
• Implementation of a graphical user interface (GUI) for browser-based access and interaction.

Main Results:

• The framework successfully collects, analyzes, and aggregates multilingual content from diverse online sources.
• Semantic analysis capabilities include effective hate speech detection and sentiment analysis.
• Evaluation through user questionnaires confirmed the framework's feasibility for non-expert users.

Conclusions:

• The proposed web framework is a viable solution for managing and analyzing large volumes of multilingual online content.
• It empowers non-expert users to engage with social media data analysis, including hate speech identification.
• The tool supports informed decision-making for journalists, researchers, and the public in understanding online discourse.