A transfer learning method from long text to short text

Abstract

The invention relates to a method of transfer learning from a long text to a short text. The method of the transfer learning from the long text to the short text is characterized by comprising step 1, obtaining data of a source domain according to tags extracted from a short text of an target domain, and extracting a seed feature set of the source domain; step 2, creating an undirected graph of social media according to a tag set of the short text of the target domain and the seed feature set of the source domain, and extracting subgraphs containing all the nodes of the tag set and the seed feature set of the target domain from the undirected graph; step 3, obtaining a new feature representation of the data of the source domain on the basis of a Laplacian Eigenmap algorithm; step 4, classifying the data of the source domain according to the new feature representation of the data of the source domain.

Description

technical field [0001] The invention relates to a long text to short text transfer learning method. Background technique [0002] With the rapid development of science and technology, Internet information is becoming more and more diverse, and short texts such as Weibo, QQ messages, and online advertisements are playing an increasingly important role in network applications. Short text data has the characteristics of few keywords, lack of context information, high-dimensional and sparse text representation, and it is difficult to fully and accurately express text features. When the target field is short text data and there is only a small amount of labeled data, it becomes a big problem to make statistics and classification of short text database information. Compared with short texts, due to the longer length of long texts, the context will carry more keywords related to the subject of the text, and the long texts appear earlier in the network, and their classification tec...

AI Technical Summary

Problems solved by technology

[0003] At present, there are some studies on transfer learning methods from long texts to short texts, but they often require source domain data related to short texts in the target domain, so that data acquisition and domain correlation measurement will be due to human Subjectivity produces some errors, which negatively affects the target task; there are also some studies that require the prior probability distribution of the data to be known before long text to short text transfer learning, which is difficult to obtain in practice

Images