Methodological Framework

The visualization presented above utilizes a multi-dimensional analysis of news content, grounded in computational linguistics and political science frameworks.

Subjectivity (Vertical Axis): The 'private state' density of each article is quantified using a lexicon-based sentiment analysis model adapted from Baccianella et al. (2010). This metric distinguishes between objective reporting (zero-degree modality) and subjective expression (opinion, speculation, and affective language), providing a continuous scale from verification-based journalism to commentary.

Political Alignment (Horizontal Axis): The lateral positioning synthesizes declared editorial affiliation with automated 'content slant' detection. Drawing upon the phrase-based bias measurement methodology proposed by Gentzkow and Shapiro (2010), our system calculates localized drift by analyzing the frequency of ideologically charged vocabulary (e.g., specific policy framing tokens). Furthermore, sentiment valence is integrated as a weighting factor; adhering to the circumplex model of affect, high-arousal negative sentiment is treated as an amplifier of ideological signaling, extending the variance beyond binary classification.

References:
Baccianella, S., Esuli, A., & Sebastiani, F. (2010). SentiWordNet 3.0: An Enhanced Lexical Resource for Sentiment Analysis and Opinion Mining.
Gentzkow, M., & Shapiro, J. M. (2010). What Drives Media Slant? Evidence from U.S. Daily Newspapers. Econometrica, 78(1), 35–71.