By Fatima Ahmad – AI Master research Vrije Universiteit Amsterdam
Stereotyping of groups or minorities in official media channels can largely and negatively influence public opinion. It is therefore important to detect stereotypes in the media and adress them. Since stereotyping is often hidden, subtle and difficult to detect in written texts, computational models are proposed. Recent studies have shown promising result using microportraits, a natural language processing method that identifies and labels certain positive, negative or stereotyping concepts, for a given inividual or group in text. In this study we use microportraits to detect stereotyping of Muslims in Norwegian conventional media. This study aims to explore the potential of the proposed NPL model to detect weak associations as stereotypical and to propose improvements to the method.