Automated identification of media bias in news articles: an interdisciplinary literature review

Abstract

Media bias, i.e., slanted news coverage, can strongly impact the public perception of the reported topics. In the social sciences, research over the past decades has developed comprehensive models to describe media bias and effective, yet often manual and thus cumbersome, methods for analysis. In contrast, in computer science fast, automated, and scalable methods are available, but few approaches systematically analyze media bias. The models used to analyze media bias in computer science tend to be simpler compared to models established in the social sciences, and do not necessarily address the most pressing substantial questions, despite technically superior approaches. Computer science research on media bias thus stands to profit from a closer integration of models for the study of media bias developed in the social sciences with automated methods from computer science. This article first establishes a shared conceptual understanding by mapping the state of the art from the social sciences to a framework, which can be targeted by approaches from computer science. Next, we investigate different forms of media bias and review how each form is analyzed in the social sciences. For each form, we then discuss methods from computer science suitable to (semi-)automate the corresponding analysis. Our review suggests that suitable, automated methods from computer science, primarily in the realm of natural language processing, are already available for each of the discussed forms of media bias, opening multiple directions for promising further research in computer science in this area.

A calibrated measure to compare fluctuations of different entities across timescales

“…Additionally, the number of sources annotated by political bias was relatively low and skewed towards the US which significantly decreased the reliability and generality of the results, although giving a more homogeneous dataset to analyze. Unfortunately, mass-scale automatic bias classification with high accuracy seems out-of-reach for now 38,39 . Moreover, more information could be extracted from similar datasets by inspection of consecutive PCs.…”

A Multidimensional Dataset Based on Crowdsourcing for Analyzing and Detecting News Bias

“…On the one hand, news articles remain the primary source to stay informed and to form opinions [8]. On the other hand, media bias has a strong impact on the individual and public perception of news topics leading to political changes [24].…”

“…Therefore, media bias issue must not be underestimated. Nearly all news consumers are affected by media bias [8]. Despite the fact that media bias analysis has a long tradition, analyzing media bias computationally with text mining methods and computational linguistics methods, as well as developing approaches to detect news bias automatically, has picked up speed only in the recent years.…”

“…Media bias detection [8] has recently gathered much attention in research communities along with other related tasks, such as fake news detection, rumor detection, and satire detection. Specifically, detecting media bias in news articles is of great importance because news articles remain the primary source to acquire information and to form opinions about events [8]. News bias detection systems can, for instance, be combined with browser plug-ins to assist online news readers with awareness of biased texts [4].…”

Orwellian-times at SemEval-2019 Task 4: A Stylistic and Content-based Classifier

“…Other work is primarily addressing related fields such as ideology detection, fake news detection. For example (Hutto et al, 2015;Hamborg et al, 2018) addresses identification and quantification of media bias. (Iyyer et al, 2014) addresses political ideology detection using recursive neural networks.…”