Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Data processing method, device and system

A data processing and data technology, applied in the field of data processing, can solve problems such as low efficiency

Pending Publication Date: 2021-05-14
ALIBABA GRP HLDG LTD
View PDF0 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] Embodiments of the present invention provide a data processing method, device and system to at least solve the technical problem of low efficiency caused by the use of man-machine asynchronous workflow in active learning in the related 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
  • Data processing method, device and system
  • Data processing method, device and system
  • Data processing method, device and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0034]According to an embodiment of the present invention, an embodiment of a data processing method is also provided. It should be noted that the steps shown in the flowcharts of the drawings can be executed in a computer system such as a set of computer-executable instructions, and, Although a logical order is shown in the flowcharts, in some cases the steps shown or described may be performed in an order different from that shown or described herein.

[0035] The method embodiment provided in Embodiment 1 of the present application may be executed in a mobile terminal, a computing device, or a similar computing device. figure 1 A hardware structure block diagram of a computing device for implementing the data processing method is shown. Such as figure 1 As shown, the computing device 10 may include one or more (shown by 102a, 102b, ..., 102n in the figure) processor 102 (the processor 102 may include but not limited to a microprocessor MCU or a programmable logic device FP...

Embodiment 2

[0098] According to an embodiment of the present invention, a data processing method is also provided, Figure 5 It is a flow chart of a data processing method according to Embodiment 2 of the present application, such as Figure 5 As shown, the method includes:

[0099] Step S51, labeling the first labeled object to obtain the first labeled object, wherein the first labeled object is the boundary data obtained after the trained learning model predicts at least one set of unlabeled data in the data pool, A data pool includes sets of data.

[0100] The above-mentioned labeling process is a process of determining the sample type of the data. In the above solution, the labeled object, that is, the first labeled object, is the boundary data collected according to the prediction result after the learning model predicts at least one set of unlabeled data in the data pool.

[0101] Specifically, the aforementioned boundary data is used to represent data for which the learning mode...

Embodiment 3

[0112] According to an embodiment of the present invention, a data processing device for implementing the data processing method in Embodiment 1 above is also provided, Figure 6 is a schematic diagram of a data processing device according to Embodiment 3 of the present application, such as Figure 6 As shown, the device 600 includes:

[0113] The first training module 602 is configured to train the learning model according to the training data in the training set.

[0114] The prediction module 604 is configured to predict at least one set of unlabeled data in the data pool using the learning model obtained in this training, and collect the first labeled object from at least one set of unlabeled data according to the prediction result, wherein the data pool are divided into groups.

[0115] The labeling module 606 is configured to label the first labeled object to obtain the first labeled object, wherein, during the process of labeling the first labeled object, the training...

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 data processing method, device and system. The method comprises the following steps: training a learning model according to training data in a training set; predicting at least one group of unlabeled data in a data pool by using a learning model obtained by this training, collecting a first labeled object from the at least one group of unlabeled data according to a prediction result, and the data pool being divided into multiple groups; marking the first marking object to obtain a first marked object, updating the training set by using a second marked object obtained by last marking in the process of marking the first marked object, and training the learning model by using the updated data set; updating a training set based on the first labeled object; and continuing to train the learning model based on the updated training set. According to the invention, the technical problem of low efficiency caused by the adoption of a man-machine asynchronous working process during active learning in the prior art is solved.

Description

technical field [0001] The present invention relates to the field of data processing, in particular to a data processing method, device and system. Background technique [0002] When establishing a data set for the deep learning model, in addition to ensuring that the amount of collected data is large enough, it is also necessary to mine boundary data through active learning to ensure that the learning model has a good non-generalization recognition ability. [0003] Conventional active learning adopts a human-machine asynchronous workflow, which includes: (A) the machine trains the machine learning model based on all the current labeled data, the learning model makes predictions on the unlabeled data pool, and collects boundary data according to certain criteria; (B) Annotators mark the boundary data returned by the machine, and then return the marked data to the machine. Usually, active learning has to go through multiple cycles of (A)-(B)-(A)-(B)-..., and each cycle of (...

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): G06N20/00
CPCG06N20/00
Inventor 刘宇光孔祥衡段曼妮
Owner ALIBABA GRP HLDG LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap 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