Topic Detection of Multi-feature Fusion in the Financial Field

Financial-related texts represent relevant trends in the financial field and influence investors' decision-making and market trends. In order to efficiently and quickly obtain topic information from the massive financial texts, topic detection technology is used to obtain the topic information of financial texts. Considering the characteristics of financial texts, this paper extracts the features of the text from the three perspectives of subject, keyword, and entity, and the time attenuation factor is used to fuse the features. Then the text is clustered based on the combined single-pass clustering and neural network clustering algorithms to obtain the subject information of the text. The experimental results show that the above method can effectively mine the subject information of financial texts.