Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

Classification method and device based on combined model, equipment and storage medium

A technology of combining models and classification methods, applied in computing models, character and pattern recognition, instruments, etc., can solve problems such as large misclassifications

Active Publication Date: 2021-03-26
TENCENT TECH (SHENZHEN) CO LTD
View PDF9 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, due to the different characteristics of each product, the classification results obtained through the unified model may exclude products that are required by users and have low scores, resulting in a large misclassification

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
  • Classification method and device based on combined model, equipment and storage medium
  • Classification method and device based on combined model, equipment and storage medium
  • Classification method and device based on combined model, equipment and storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0031] figure 1 A flow chart of a classification method based on a combined model is provided for the present application embodiment. The execution body of the method can be a desktop computer, a notebook, a mobile phone, a vehicle terminal, and other smart electronic devices, or a processor in an electronic device. This application does not do limitations, such as figure 1 As shown, the method includes the following steps:

[0032] S110: Gets M group training samples and N group prediction data, M and N are positive integers.

[0033] S120: Training the M models by M group training samples, respectively, to obtain M target models.

[0034] S130: For any target model in the M target model, according to the target model, the target model corresponding to the first training sample, the classification threshold corresponding to the first training sample, to obtain the first training samples in the first training sample A category label. The second classification label included in ea...

Embodiment 2

[0066] Optionally, in this application, the classification threshold can be fixed or dynamically calculated, and this application does not limit this.

[0067] The method of dynamically calculating classification thresholds will be explained below. It should be understood that the following mainly describes the method of dynamically calculated classification thresholds involved in the model training process, and the method of dynamic calculation classification thresholds involved in the model test process and prediction process can refer to the dynamic calculation classification during the training process. The method of threshold, this application is not described again.

[0068] Optionally, the classification threshold corresponding to the first training samples can be determined by the formula:

[0069]

[0070] among them, Represents the classification threshold corresponding to the first training sample group, Represents the number of training samples in the first trainin...

Embodiment 3

[0074] Optionally, after the completion of the M model training, the electronic device can also test the target model after m training after M-Test samples. Based on this, when the M target model is tested, the electronic device is based on the difference between the respective classification tags corresponding to the M target model and the M target model, and obtains a combined model. When any model test in the M target model fails, the electronic device re-acquires the training sample group corresponding to the test failure, to continue to train the model, and re-acquire the test sample group corresponding to the model, continue This model is tested until all model tests are successful.

[0075] Optionally, after the electronic device acquires the combined model, the combined model can be tested, and the prediction process is performed when the combined model is tested. When the combined model test fails, the electronic device re-acquires the M group training sample and the N gr...

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 classification method and device based on a combined model, equipment and a storage medium, and can be applied to the classification fields of social contact, games, multimedia, vehicle connection, driving, shopping and the like. The method comprises the steps of obtaining m groups of training samples and n groups of prediction data; training the m models through the m groups of training samples to obtain m target models; obtaining a first classification label of each training sample in the first training sample group according to the target model, the first training sample group corresponding to the target model and the classification threshold of the first training sample group; obtaining a classification label variance corresponding to the target model accordingto the first classification label of each training sample and the second classification label included in each training sample; obtaining a combined model according to the m target models and the classification label variances corresponding to the m target models; and according to the n groups of prediction data, the combination model and the classification thresholds corresponding to the n groups of prediction data, obtaining a classification label of each prediction data so as to improve the accuracy of the classification method.

Description

Technical field [0001] The present application embodiments relate to the technical field of artificial intelligence (AI), and more particularly to a classification method, apparatus, apparatus, and storage medium based on a combined model. Background technique [0002] Many scenarios are related to classification methods, for example, in a scenario recommended to users involving classification methods, that is, by classification methods to determine if users are interested in each item, if they are determined to interested in a product, The product is recommended to the user, otherwise, the item is not recommended to the user. [0003] The above classification method is based on the unified model, for example, for multiple products, unified models to determine if the user is interested in each item. This unified model is obtained by training sample training in each commodity, which includes: user features, commodity features, and classified labels. However, since the characterist...

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
IPC IPC(8): G06K9/62G06N20/00G06Q30/02G06Q30/06
CPCG06N20/00G06Q30/0201G06Q30/0631G06F18/241
Inventor 钟子宏
Owner TENCENT TECH (SHENZHEN) 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
Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com