Analysis of International Media Portraits via Knowledge Distillation and Domain Adaptive Models

Global media with international influence play an extremely important role in shaping international public opinion platforms related to China. Studying international public opinion orientation plays an important role in building a positive national image, improving the international public opinion environment, and resisting various ideological infiltration. In this study, we take the 20th National Congress of the Communist Party of China as a typical case to construct an international media portrait and conduct a fine-grained analysis of media attributes. This paper uses the LDA topic model to explore the hot spots of international media news and uses knowledge distillation and adversarial domain adaptive methods to efficiently and accurately analyze the public opinion tendency of international China-related reports. The research results show that some Western media still have a prejudice against China, and the construction of an international media portrait is of great significance and value for a timely, comprehensive, and accurate grasp of the international public opinion situation.