User tag extraction method and device, equipment and medium

Abstract

The invention relates to the field of artificial intelligence, and provides a user tag extraction method and device, equipment and a medium, which can recall all possible text expressions through text enhancement, and then screen the text obtained after enhancement through sensitive word recall so as to ensure that keywords are accurately positioned as many as possible for subsequent model training. The model training effect is improved. A user tag extraction model is obtained by combining sensitive word recall and deep learning training; meanwhile, the high adaptability of sensitive word recognition in a public opinion scene and the high accuracy of a deep learning algorithm for text sentiment tone judgment are fused. A large number of recalled public opinion related texts are recognized through sensitive words, and the emotional positivity and negativity of the recalled text are judged through deep learning, so that the accuracy of tag identification can be effectively improved. In addition, the invention also relates to a block chain technology, and the user tag extraction model can be stored in a block chain node.

Description

technical field [0001] The present invention relates to the technical field of artificial intelligence, in particular to a method, device, equipment and medium for extracting user tags. Background technique [0002] With the development of the economy, the country will continue to promote strict supervision of the financial industry. Under the trend of continuous strict supervision, major financial groups need more advanced and effective risk management methods, and are committed to providing customers with better products and services. Therefore, it is necessary to accurately identify the emotional state of customers during the service process. [0003] However, when a public opinion event occurs, it is often the agents who come into contact first. On the one hand, due to the lack of awareness of macro control by agents, it is easy to underestimate the influence of events, resulting in concealment of public opinion events; on the other hand, although the management has a s...

AI Technical Summary

Problems solved by technology

[0005] (1) It is difficult for the deep learning algorithm to identify the interactive long text between customer service and users, which not only requires a large amount of computing resources, but also needs to mark a large number of non-public opinion texts to recall enough texts due to the extremely low frequency of public opinion related texts. The sample text of public opinion has a huge labor cost;

[0006] (2) The generation of public opinion is mostly based on only one keyword, such as swear words. The characteristics of a word of swear words in a long sentence are not obvious, and it is difficult to be recognized

Images