Unlock instant, AI-driven research and patent intelligence for your innovation.

Chinese text classification method based on multi-input attention network

A text classification and attention technology, applied in text database clustering/classification, neural learning methods, biological neural network models, etc., can solve problems such as unsatisfactory, no interpretability, and incomplete utilization of language features. Achieve high reliability and high classification accuracy

Active Publication Date: 2022-07-15
CENT SOUTH UNIV
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

(2) Incomplete utilization of language features
(3) The results are not interpretable
At present, the existing research is not satisfactory

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Chinese text classification method based on multi-input attention network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0042] like figure 1 Shown is a schematic flow chart of the method of the method of the present invention: this multi-input attention network-based Chinese text classification method provided by the present invention includes the following steps:

[0043] S1. Obtain Chinese text data;

[0044] S2. According to the Chinese text data obtained in step S1, establish a corresponding language model; specifically, the following steps are used to establish a language model:

[0045] A. Segment the acquired Chinese text data and remove stop words;

[0046] B. the Chinese text that step A obtains is converted into corresponding pinyin text;

[0047] C. the Chinese text that step A obtains and the pinyin text that step B obtains are counted respectively, obtain Chinese text statistical data and pinyin text statistical data;

[0048] D. The Chinese text statistics and pinyin text statistics obtained in step C are trained to obtain matrix data based on word vectors;

[0049] In the spe...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a Chinese text classification method based on a multi-input attention network, including acquiring Chinese text data; establishing a corresponding language model; establishing a pinyin encoder and a text encoder; combining the pinyin encoder and the text encoder to obtain a preliminary The new multi-input attention network model is optimized and the final new multi-input attention network model is obtained; the final new multi-input attention network model is used to classify the input Chinese text to obtain the final classification result. The Chinese text classification method based on the multi-input attention network provided by the present invention adopts a novel multi-input attention network structure to realize the classification of Chinese text in natural language, so the method of the present invention has high reliability and accurate classification High rate and relatively simple.

Description

technical field [0001] The invention belongs to the field of natural language data processing, in particular to a Chinese text classification method based on a multi-input attention network. Background technique [0002] Nowadays, artificial intelligence has been widely used in various fields, especially in the field of natural language processing. Artificial intelligence systems have achieved remarkable results in text classification, text generation, machine translation, and machine reading. In the field of natural language processing, artificial intelligence has also developed greatly, and its fast and accurate summarization, classification, translation and generation are unmatched by humans. Using artificial intelligence natural language processing systems, the processing time of language information that used to take days or even weeks can be greatly shortened. This is conducive to quickly processing various language information, saving human resources, reducing relate...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06F16/35G06F40/289G06F40/126G06K9/62G06N3/04G06N3/08
CPCG06F16/35G06N3/08G06N3/045G06F18/2415
Inventor 仇俊豪施荣华张帆
Owner CENT SOUTH UNIV