Deep Learning and Multi-Feature Fusion based New Media Communication Event Clustering Study

With the vigorous development of new media, the number and variability of communication events are increasing, and the traditional event clustering methods can no longer meet the demand for fast and accurate clustering of large-scale and multi-featured communication events. To address this problem, this paper proposes a new method. Firstly, we use deep learning algorithm to extract features from propagation events, which effectively improves the feature characterization ability by automatically learning abstract features from data. At the same time, we adopt a multi-feature fusion approach to fuse different types of features, including text features, image features, video features, etc., in order to reflect the diversity of communication events more comprehensively. Finally, the fused features are clustered using a clustering algorithm to achieve automatic classification and generalization of communication events. Through experimental verification, our method outperforms traditional clustering methods in terms of accuracy and efficiency. The method provides a new idea and solution for new media communication event clustering, which is of great significance for improving the efficiency and accuracy of public opinion monitoring and management.