Multi-label corpus text classification method based on semi-supervised learning

Abstract

The invention discloses a multi-label corpus text classification method based on semi-supervised learning, and the method comprises the following steps: carrying out the semi-supervised learning basedon a multi-label corpus text, and obtaining a classification strategy knowledge base; preprocessing the corpus text to be classified; classifying the corpus classification texts, and determining a first text content identifier set; determining a first text content set in the preset training data set, and in the first text content set, selecting text contents corresponding to N candidate categories according to the certain number of candidate categories to determine a second text content set; and determining a target category of the to-be-classified text according to the similarity between thetext feature words and each piece of text content in the second text set. The method has the advantages that the calculation complexity and the calculation amount are reduced, and the efficiency of the text class is improved.

Description

technical field [0001] The invention relates to the field of corpus text classification, in particular to a semi-supervised learning-based multi-label corpus text classification method. Background technique [0002] Text classification is an important content of text mining, which refers to determining a category for each document in a document collection according to a predefined subject category. Classifying documents through an automatic text classification system can help people better find the information and knowledge they need. Classification is seen as the most basic form of cognition of information. [0003] With the rapid growth of text information, especially the surge of online text information on the Internet, automatic text classification has become a key technology for processing and organizing large amounts of document data. Now, text classification is widely used in various fields. For example, on the Internet platform, the server can classify the text in...

AI Technical Summary

Problems solved by technology

[0004] In the existing methods for classifying text, as the amount of information becomes more and more abundant, people will have higher and higher requirements for the accuracy rate and recall rate of content search, and the number of samples contained in the training set is also very large. Calculating the similarity with each sample in the training set by traversal requires a lot of performance of the server, and the calculation speed is slow

As a result, the effective resources of the server are occupied in large quantities, and the calculation time is too long, so it takes a lot of time to answer or push relevant information to the user

Images