University of Southern California (USC) computer scientists have developed an artificial intelligence tool to automatically detect bias in news through a combination of natural-language processing and moral foundation theory.
Rather than just searching for keywords, the algorithm considers more complex patterns in the contextual application of language to evoke key themes or frames.
The algorithm determined content posted on left-leaning sites highlighted fairness and equity, while material on right-leaning sites emphasized moral purity under fear of contamination and appeals to law and order.
USC's Kristina Lerman said, "Our ultimate goal was to grade news to create a news bias 'nutrition label' for when you are consuming news."
From USC Viterbi School of Engineering
View Full Article
Abstracts Copyright © 2020 SmithBucklin, Washington, DC, USA
No entries found