Recommendation method and system based on unsupervised learning

An unsupervised learning and recommendation method technology, applied in neural learning methods, special data processing applications, instruments, etc., can solve problems such as a lot of manpower, reduce resources and alleviate insufficient labeling resources.

Pending Publication Date: 2022-01-07
新华智云科技有限公司
View PDF0 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0008] The purpose of the present invention is to provide a recommendation method and system based on unsupervised learning to solve the techn

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
  • Recommendation method and system based on unsupervised learning

Examples

Experimental program
Comparison scheme
Effect test

Example Embodiment

[0045] Example 1:

[0046] like figure 1 As shown, this embodiment includes a recommended method based on an undo-monitoring learning, including the following steps: acquiring the first training data, the first training data inputs the BERT model for the first time, obtain the second training data, the first training data second Enter the Bert model to get a third training data;

[0047] The second training data and the third training data constitute a positive sample pair and negative sample pair, the active pair and negative sample input LOSS function, obtain the loss value; if the loss value is within the preset threshold, if not, repeat operation acquisition first Training data, the first training data enters the BERT model for the first time and the following steps; if yes, the precision BERT model is obtained;

[0048] Get user data, accurate the processing-after input to accurate the Bert model, calculate the similarity score of the user data and the data to be recommended,...

Example Embodiment

[0058] Example 2:

[0059] This embodiment includes a recommendation system based on an undisperate learning, including:

[0060] The first training module is used to obtain the first training data. The first training data enters the BERT model for the first time to obtain the second training data, the first training data inputs the BERT model for the second training data;

[0061] The second training module, for the second training data and the third training data constitute a positive sample pair and negative sample pair, the active pair and negative sample input LOSS function, resulting in the loss value.

[0062] The determination module is used to determine whether the loss value is within the preset threshold, if, repeatedly runs the first training data, the first training data enters the BERT model for the first time and the steps; if yes, the precision BERT model is obtained;

[0063] Recommended module, obtain user data, accurate the processing-proof user data into the BE...

Example Embodiment

[0065] Example 3:

[0066] A computer readable storage medium, computer readable storage medium stores a computer instruction, and the computer instruction is implemented in the process of the method in which the method in Example 1 is performed.

[0067] Those skilled in the art will appreciate that embodiments of the invention may be provided as a method, apparatus, or computer program product. Accordingly, the present invention can employ a full hardware embodiment, a fully software embodiment, or in combination with software and hardware aspects. Moreover, the present invention can employ a computer program product that includes a computer available storage medium (including but not limited to disk memory, CD-ROM, optical memory, etc.) implemented in one or more computers.

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 recommendation method and system based on unsupervised learning. The method comprises the following steps: obtaining first training data, inputting the first training data into a bert model for the first time to obtain second training data, and inputting the first training data into the bert model for the second time to obtain third training data; forming a positive sample pair and a negative sample pair by the second training data and the third training data, and inputting the positive sample pair and the negative sample pair into a loss function to obtain a loss value; judging whether the loss value is within a preset threshold value or not, and if yes, obtaining an accurate bert model; and obtaining user data, inputting the processed user data into a precise bert model, calculating a similarity score of the user data and to-be-recommended data, and recommending the user according to the similarity score. The invention has the beneficial effects that the problems of multi-layer feature representation and one-word polysemy of words can be better solved, resources needing manual annotation are reduced, the problem of insufficient annotation resources can be well relieved, and meanwhile the recommendation speed can be greatly increased.

Description

technical field [0001] The invention belongs to the technical field of recommendation methods, and more specifically, the invention relates to a recommendation method and system based on unsupervised learning. Background technique [0002] In recommendation system projects, such as the recommendation of film and television works, text similarity is usually used to recommend similar items. Text similarity is a commonly used recommendation algorithm. The application of text similarity in recommender system projects can be attributed to calculating the similarity score for the profile text, sorting by the similarity score, and recommending according to the sorting results. In the recommendation projects of industrial scenarios, we often face the problem of unlabeled data, and we also need to consider the performance of the recommendation algorithm. The current technical solutions for the recommendation projects of industrial scenarios have the following defects: [0003] First...

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): G06F16/9035G06K9/62G06N3/08
CPCG06F16/9035G06N3/088G06F18/22G06F18/214
Inventor 郭鑫润
Owner 新华智云科技有限公司
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