Rumor Detection Based on Generalized Text Feature

This paper proposes a rumor detection method based on topic classification and content understanding in health. It can extract features from different scales of sub-datasets, comprehensively consider the correlation and difference in different topics, and combine the part of speech and word meaning to expand the model's ability to understand the text and enhance the ability to detect the rumor traps spread maliciously. The experimental results show that the method achieves good results in the field of health data sets, improving the results in new indicators such as content understanding, and have more vital generalization ability.