Progress in Cross Language Text Sentiment Analysis Based on Deep Learning

With the acceleration of globalization, cross-language information processing technology is getting more and more attention. Among them, cross-lingual text sentiment analysis, as an important branch in the field of sentiment computing, understands and analyzes the user's emotional state in a multilingual environment, which is of great significance for improving user experience and optimizing decision making. In recent years, cross-lingual sentiment analysis methods based on deep learning have made significant progress. This paper summarizes the recent research progress in this field. Firstly, we introduce the research background and challenges of cross-lingual text sentiment analysis, including language differences, domain differences, data scarcity and other issues. Then, we focus on the application of deep learning-based methods, such as cross-language word embedding, adversarial transfer learning, meta-learning and other techniques in cross-lingual sentiment analysis. It is found that these methods can effectively capture the semantic and emotional links between different languages and significantly improve the accuracy of cross-lingual sentiment analysis. This paper discusses future research directions in this area, including fusing multimodal information, enhancing the generalization ability of models, and combining deep learning techniques with traditional knowledge-based methods.