Data information classification method and device

A technology of data information and classification methods, which is applied in the field of information processing, can solve problems such as the difficulty in determining the decision boundary of Bayesian models, and the inability of high-dimensional discriminant model vector representations to represent semantic information, etc., achieving the effect of low cost and simple and efficient methods

Active Publication Date: 2021-02-05
SHANGHAI XIAOI ROBOT TECH CO LTD
View PDF6 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Bayes, KNN is a Bayesian model based on statistics. The main problem of the high-dimensional discriminant model is that the semantic information of the complete text cannot be represented in the vector representation, and the decision boundary determination of the Bayesian model is very difficult.

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
  • Data information classification method and device
  • Data information classification method and device
  • Data information classification method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0058] Such as figure 1 As shown, this embodiment provides a method for classifying data information, including the following steps:

[0059] Step S10, acquiring text information to be classified;

[0060] Step S20, performing vectorization processing on the text information to obtain a text vector corresponding to the text information;

[0061] Step S30, performing fusion processing on the text vectors to obtain representation features of multiple aspects of the text information;

[0062] Step S40, performing a global average pooling process on the feature vectors included in the representation features to obtain aggregation information corresponding to each of the feature vectors;

[0063] Step S50, using two fully-connected networks to screen the aggregated information to obtain screening parameters corresponding to each of the feature vectors;

[0064] Step S60, judging whether the feature vector is a noise feature according to the screening parameters, if so, masking t...

Embodiment 2

[0120] Such as figure 2 As shown, this embodiment provides a data information classification device, including:

[0121] Input module 100, for obtaining the text information to be classified;

[0122] A vectorization module 200, configured to perform vectorization processing on the text information to obtain a text vector corresponding to the text information;

[0123] A fusion module 300, configured to perform fusion processing on the text vectors to obtain representation features of multiple aspects of the text information;

[0124] The pooling module 400 is configured to perform global average pooling processing on the feature vectors included in the representation features, to obtain aggregation information corresponding to each of the feature vectors;

[0125] A screening module 500, configured to use two fully connected networks to screen the aggregated information to obtain screening parameters corresponding to each of the feature vectors;

[0126] The denoising mod...

Embodiment 3

[0152] Such as image 3 As shown, the present embodiment provides an electronic device 90, comprising: one or more processors 91 and a memory 92; and computer program instructions stored in the memory 92, which when executed by the processor 91 cause the processor 91 executes each step of the data information classification method described in the first embodiment.

[0153] It should be noted that the data information classification apparatus according to the embodiment of the present application may be integrated into the electronic device 90 as a software module and / or hardware module, in other words, the electronic device 90 may include the data information classification apparatus. For example, the data information classification device may be a software module in the operating system of the electronic device 90, or may be an application program developed for it; of course, the data information classification device may also be the electronic device 90 One of many hardwar...

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 provides a data information classification method and device. The method comprises steps of obtaining the to-be-classified text information; sequentially performing vectorization processing, fusion processing and global average pooling processing on the text information to obtain the aggregated information corresponding to each feature vector; performing screening processing on the aggregation information by using two fully connected networks to obtain a screening parameter corresponding to each feature vector; judging whether the feature vectors are noise features or not according to the screening parameters, if yes, shielding the feature vectors, and updating the remaining feature vectors to obtain updated feature vectors; updating representation features of multiple aspects according to the updated feature vectors; dimension reduction processing being performed on the updated representation features of the multiple aspects so that a target feature can be obtained; andobtaining the classification information of the text information according to the target features. According to the method, the classification purpose can be achieved more comprehensively and accurately, and the method is simple, efficient and low in cost.

Description

technical field [0001] The invention relates to the technical field of information processing, in particular to a data information classification method, a data information classification device, a storage medium and electronic equipment. Background technique [0002] With the rapid development of the information age, the information resources on the Internet are becoming more and more abundant, the scale of information data is becoming larger and larger, and the forms of expression are becoming more and more diverse. However, for massive information and data resources, most of them can only be understood by humans, and it is still very difficult for machines to understand this information, especially the huge amount of text data, and natural language understanding has always been very popular. research field. [0003] In the process of natural language processing, text classification is particularly important as the basis for applications such as content classification, se...

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 Applications(China)
IPC IPC(8): G06F16/35G06N3/04
CPCG06F16/355G06N3/045Y02D10/00
Inventor 陈成才
Owner SHANGHAI XIAOI ROBOT TECH CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap