Multi-label Classification Algorithm for Financial Texts Based on Deep Learning

The multi-label classification algorithm for financial texts can implement information retrieval from vast financial data according to user needs. To further enhance the ability to identify financial text labels, this paper proposes a deep learning-based multi-label classification algorithm for financial texts. Deep learning captures both local and global graph structures through deep neural networks, allowing for the modeling of complex relationships between nodes. By modeling the dependencies among labels, knowledge transfer between labels can be realized, which is the key to building a model with strong generalization capability. To enhance the correlation information between labels, we adopt a dual-gated recurrent neural network combined with attention mechanisms to generate characteristic representations of news texts under different labels, and model the complex dependencies between labels through graph neural networks. The experimental results on real datasets demonstrate that the model significantly improves the generalization ability of label correlation modeling, especially for tail labels. Compared to CAML, BIGRULWAN, and ZACNN algorithms, this method achieves the highest improvement in macro-F1 score for all and tail labels by 3.1% and 6.9%, respectively.