A Free Press, If You Can Keep It : What Natural Language Processing Reveals About Freedom of the Press in Hong Kong

This Brief introduces a novel research approach to investigate freedom of the press in Hong Kong. The authors pair computational analyses from the field of natural language processing with qualitative content analysis of patterns of journalistic practice in volatile political settings. Together, these shed light on the evolution of press freedom in Hong Kong since its return to Chinese sovereignty. Providing an interdisciplinary perspective, the Brief will appeal to a wide range of readers with interests in computational social science, public policy, political sciences as well as policy-makers, think tanks, and practitioners who focus on the China-Hong Kong nexus. Acknowledgments Contents Chapter 1: Introduction Chapter 2: Methodological Approach and Data 2.1 Methodological Approach 2.2 The Hong Kong Protest News Dataset Chapter 3: Non-rhetorical Tactics 3.1 Volume and Timing of Articles on Hong Kong Protests 3.2 Does News Coverage Granger-Cause Protest Size? Chapter 4: Rhetorical Tactics 4.1 The Protest Paradigm: Framing the Protests 4.2 BERT-MALLET Vicinato Plots 4.3 The Lexicon Used to Portray the Hong Kong Protests 4.3.1 Democracy and Freedom 4.3.2 Protests 4.4 Differences in How Hong Kong – And Western-Based Newspapers Portray the Protests 4.4.1 Sentiment Analysis 4.4.2 Comparing Embedding Neighborhoods 4.4.3 Does Coverage Differ Before and After the Outbreak of Protests in June 2019? 4.4.4 News Headline Analysis Chapter 5: Conclusions References