Keyword extraction method, device, equipment and medium based on information entropy

Abstract

The present invention relates to the field of artificial intelligence, and provides a keyword extraction method, device, equipment and medium based on information entropy, which can preprocess the received label text, and reduce the redundancy of characters while standardizing the text. Improve the speed and effect of data processing, and use the inverse information entropy vector. On the one hand, it can replicate the effect of TF-IDF in the case of high data quality. The keywords of the distinguishing ability effectively correct the failure of the traditional TF-IDF’s category distinguishing ability, thereby avoiding misjudgment, and have high interpretability. The target keywords are extracted according to the word weight matrix to realize automatic and accurate keyword identification. extract. In addition, the present invention also relates to block chain technology, and target keywords can be stored in block chain nodes.

Description

technical field [0001] The present invention relates to the technical field of artificial intelligence, in particular to a keyword extraction method, device, equipment and medium based on information entropy. Background technique [0002] In the field of artificial intelligence, keyword extraction plays an important role. For example, in the process of enterprise employee training, more and more links have gradually shifted from offline to online. Taking the professional training of life insurance agents as an example, if the training is to be practical and effective, so that agents have the opportunity to apply what they have learned to the actual life insurance sales, customer retention and other scenarios, the training and production departments must differentiate between the crowd and the training. In-depth research on chemistry. The traditional method mainly revolves around conducting regular communication with the business department to collect the appeals and points ...

AI Technical Summary

Problems solved by technology

[0003] However, because the traditional keyword extraction technology only considers the number of documents in which a certain word appears, it is very easy to fail in a scene with a similar text language environment, and it is difficult to play the role of discovering differentiated keywords. Words only appear once in other categories of people, and because the word appears in every document, the word will be treated as a "common word", thereby reducing the weight of the word in the vocabulary of the category to which it belongs, resulting in recognition error

Images