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

A text classification method and device

A text classification and text technology, applied in the computer field, can solve the problems of inaccurate text classification, reduce the accuracy and efficiency of text classification, etc., and achieve the effect of improving interpretability, accuracy, accuracy and efficiency

Active Publication Date: 2020-09-11
TENCENT TECH (SHENZHEN) CO LTD
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] Embodiments of the present invention provide a text classification method and device to solve the problem of inaccurate text classification and reduce the accuracy and efficiency of text classification in the prior art

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
  • A text classification method and device
  • A text classification method and device
  • A text classification method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment approach

[0059] Specifically, for determining the connection probability between corresponding every two adjacent participles, the embodiment of the present invention also provides a possible implementation manner, which specifically includes:

[0060] S4.1. Input the weights of each relationship into the first activation function, and transform the weights of each relationship to a preset value range through the first activation function.

[0061] For example, for the Tanh function, the corresponding preset value range is (-1,1), and the value of the relationship weight can be mapped to (-1,1).

[0062] S4.2. Input the transformed relationship weights into the second activation function, and if it is determined that the value of any transformed relationship weight is less than 0, adjust the transformed relationship weights with values ​​less than 0 to 0 , if it is determined that the value of the transformed relationship weight is not less than 0, no adjustment is made.

[0063] That...

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 relates to the technical field of computers, in particular to a text classification method and device, and the method comprises the steps: obtaining feature vectors of all segmented words in a word segmentation result of a text input by a current-layer semantic module; Determining a connection probability between adjacent segmented words of each segmented word, and recording an indexposition with the connection probability of 0; Determining a new word segmentation result of the text based on a recurrent neural network according to the determined connection probabilities and therecorded index positions; Taking a new word segmentation result of the text as the input of a next semantic module; Or, when it is confirmed that the current layer is the preset layer, obtaining the category of the text according to the new word segmentation result and the classification model, so that through the multi-layer semantic module, the accuracy of final text division is improved, and then the accuracy and efficiency of text classification are improved.

Description

technical field [0001] The invention relates to the field of computer technology, in particular to a text classification method and device. Background technique [0002] At present, when performing text understanding related tasks such as text classification, it is usually to first segment the text or word segmentation, and train the corresponding model, and then classify the text or perform related tasks based on the trained model. However, the existing The method in the technology is mainly to determine the word segmentation result according to a certain length of word segmentation or word segmentation method, and provide input features for the model. In this text segmentation method, the word segmentation results obtained may not be accurate, resulting in the input features of the model. The introduction of noise features reduces the accuracy and efficiency of model training and prediction, increases the learning difficulty of the model itself, and makes the accuracy of t...

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): G06F40/289G06F40/30G06K9/62G06N3/04G06N3/08
Inventor 王兴光牛成
Owner TENCENT TECH (SHENZHEN) CO LTD