Combination-type text classification method based on deep learning

Abstract

The invention relates to a combination-type text classification method based on deep learning. The method includes the following steps: preprocessing to-be-classified text; using the preprocessed textto obtain feature extraction results, which correspond to all deep learning models, through the different depth learning models; combining all the feature extraction results through a fully connectedlayer to obtain a combined feature vector; and using the combined feature vector as input to carry out classification recognition to obtain a class to which the text belongs. Compared with the priorart, the method has the advantages of lower requirements on training-corpus quality, high transplantability and expandability, a still better effect in a case of more classification classes and the like.

Description

technical field [0001] The invention relates to the field of natural language processing, in particular to a combined text classification method based on deep learning. Background technique [0002] With the rapid development of the Internet in the world, the Internet has gradually become one of the main carriers for people to obtain and disseminate information. Based on this point, when the Internet has become an important information dissemination tool and there are massive text data, it is of great significance to classify text based on the field of natural language processing and combined with related fields. The current text classification technology can be mainly divided into two types, one is text classification technology based on machine learning, and the other is text classification technology based on deep learning. At present, text classification technology based on deep learning has gradually become the mainstream of text classification. [0003] However, in t...

AI Technical Summary

Problems solved by technology

[0003] However, in the face of large-scale and multi-category text classification, the text classification technology based on machine learning has the characteristics of sparsity and irregularity in texts from the Internet, which makes it difficult to use these traditional classification algorithms to classify texts. The accuracy rate is generally low when

At present, text classification methods based on deep learning mostly use a single deep learning model for classification, which has high requirements for training corpus and poor portability and scalability, and due to the limitation of features extracted by a single deep learning model, resulting in many classification categories. The classification effect is poor in the case of

[0004] The paper "Research on Text Classification Based on Deep Learning Mixed Model" discloses a text classification method based on a mixed model. The hybrid model, and then use softmax regression as the classification layer to obtain text classification results. Although this method is for mixing two deep learning models, this kind of mixing is aimed at a specific deep learning model. At the same time, this kind of mixing is to be The classification text is obtained by classifying the two-layer deep learning model in turn, that is, the two deep learning models are connected in series. The result obtained by the series can actually be regarded as a deep learning model, and it does not fundamentally solve the problem of a single deep learning model. The features extracted by the model have limitations, so in the case of a large number of classification categories, the classification results obtained are still flawed

Images