Requirements Modelling For Hate Speech Detection and Analysis

The widespread use of the internet and social media has facilitated the rapid spread of information, but also of hostile narratives, which are strategic communication efforts intended to manipulate, divide, or incite harm, often taking the form of disinformation, hate speech, or deepfakes. Stakeholders, specifically Security Concerned Enterprises, require tools that are not only technically robust but also context-aware and socially responsive to support timely, accountable, and ethical decision-making. This paper presents a structured methodology that supports forward and backward traceability of requirements from fuzzy user needs expressed in natural language to system specification. The underlying theory is supported by examples in Hate Speech.