Text classification method and text classification system

Abstract

The embodiment of the invention provides a text classification method and a text classification system, and the text classification method comprises the steps: obtaining a target text, and employing aconvolution network to encode the target text into a first text vector; dividing a source text in each text library into a plurality of categories by adopting a clustering model, wherein each category is represented by a second text vector corresponding to the clustering center of the category; calculating the distances between the first text vector and the second text vectors respectively, withthe text library where the clustering centers corresponding to the nearest K second text vectors are located serves as a candidate text library of the target text through a sorting model, wherein K isa positive integer. A target text is encoded into the first text vector, distance calculation is carried out on the first text vector and the second text vectors corresponding to the multiple clustering centers in each text library, the K text libraries closest to one another serve as candidate text libraries of the target vector, the target text can be classified rapidly and efficiently, and thetext classification accuracy is improved.

Description

technical field [0001] The invention relates to the technical field of computers, in particular to a text classification method and a text classification system. Background technique [0002] With the development of "big data" technology, more and more fields are beginning to use computers for text matching and classification, and with the rapid growth of text data sets, there are more and more algorithms for text classification, and the computational complexity is also increasing. Higher and higher. [0003] Commonly used text classification methods are based on association rules or keyword-based classification. These methods are based on keywords contained in the target text or according to the association relationship. When a target text contains a certain keyword , it can be better classified. If the target text does not contain pre-selected keywords, the text cannot be classified. Such a classification method causes many target texts to be unable to be matched to the a...

AI Technical Summary

Problems solved by technology

[0002] With the development of "big data" technology, more and more fields are beginning to use computers for text matching and classification, and with the rapid growth of text data sets, there are more and more algorithms for text classification, and the computational complexity is also increasing. Higher and higher

[0003] Commonly used text classification methods are based on association rules or keyword-based classification. These methods are based on keywords contained in the target text or according to the association relationship. When a target text contains a certain keyword , it can be better classified. If the target text does not contain pre-selected keywords, the text cannot be classified. Such a classification method causes many target texts to be unable to be matched to the appropriate classification, and the keyword-based The recommendation method is relatively rough, the semantic information is not well considered, and the classification results are not accurate

[0004] Therefore, the inventor believes that the above-mentioned text classification method has great limitations. Classifying the target text based on keywords has the problem of time-consuming and labor-intensive searching and inaccurate classification.

Images