Joint recommendation system based on joint learning framework

A recommendation system and framework technology, applied in the field of joint recommendation system based on the joint learning framework, can solve the problems of manpower and material resources

Pending Publication Date: 2022-07-19
ENNEW DIGITAL TECH CO LTD
View PDF1 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] However, the specific point of the text in the health field is that it is highly professional. If the model is directly used to train the NER model on the health report information in the health field, then labeling samples will consume a lot of manpower and material resources.

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
  • Joint recommendation system based on joint learning framework
  • Joint recommendation system based on joint learning framework
  • Joint recommendation system based on joint learning framework

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0035] see Figure 1 to Figure 2, a joint recommendation system based on a joint learning framework, including target domain input 1, main LSTM layer 2, main hidden layer 3, merge layer 4, main activation function layer 5, source domain input 6, auxiliary LSTM layer 7, The auxiliary hidden layer 8, the joint hidden layer 9, the auxiliary activation function layer 10, the auxiliary task output terminal 11 and the main task output terminal 12, the target field input terminal 1 receives the relevant data predicted by the target field, and generates the corresponding main task according to the relevant data. task, and send the main task to the main LSTM layer 2, the main LSTM layer 2 receives the main task, and makes the main task enter the LSTM network to participate in the judgment of the word segmentation NER model. At the same time, the main LSTM layer 2 is based on the target domain. Generate the main hidden layer 3, the target domain input 1 is connected to the target domain...

Embodiment 2

[0043] see Figure 1 to Figure 2 , a joint recommendation system based on a joint learning framework, including target domain input 1, main LSTM layer 2, main hidden layer 3, merge layer 4, main activation function layer 5, source domain input 6, auxiliary LSTM layer 7, The auxiliary hidden layer 8, the joint hidden layer 9, the auxiliary activation function layer 10, the auxiliary task output terminal 11 and the main task output terminal 12, the target field input terminal 1 receives the relevant data predicted by the target field, and generates the corresponding main task according to the relevant data. task, and send the main task to the main LSTM layer 2, the main LSTM layer 2 receives the main task, and makes the main task enter the LSTM network to participate in the judgment of the word segmentation NER model. At the same time, the main LSTM layer 2 is based on the target domain. Generate the main hidden layer 3, the target domain input 1 is connected to the target domai...

Embodiment 3

[0051] see Figure 1 to Figure 2 , a joint recommendation system based on a joint learning framework, including target domain input 1, main LSTM layer 2, main hidden layer 3, merge layer 4, main activation function layer 5, source domain input 6, auxiliary LSTM layer 7, The auxiliary hidden layer 8, the joint hidden layer 9, the auxiliary activation function layer 10, the auxiliary task output terminal 11 and the main task output terminal 12, the target field input terminal 1 receives the relevant data predicted by the target field, and generates the corresponding main task according to the relevant data. task, and send the main task to the main LSTM layer 2, the main LSTM layer 2 receives the main task, and makes the main task enter the LSTM network to participate in the judgment of the word segmentation NER model. At the same time, the main LSTM layer 2 is based on the target domain. Generate the main hidden layer 3, the target domain input 1 is connected to the target domai...

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 joint recommendation system based on a joint learning framework, which belongs to the technical field of joint recommendation systems and comprises a target domain input end, a main LSTM (Long Short Term Memory) layer, a main hidden layer, a merging layer, a main activation function layer, a source domain input end, an auxiliary LSTM layer, an auxiliary hidden layer, a joint hidden layer, an auxiliary activation function layer, an auxiliary task output end and a main task output end. The target domain input end receives related data predicted by a target domain, generates a corresponding main task according to the related data and sends the main task to the main LSTM layer, the main LSTM layer receives the main task and enables the main task to enter an LSTM network to participate in judgment of a word segmentation NER model, and meanwhile, the main LSTM layer generates a main hidden layer on the basis of the target domain. Through a joint learning method, the source field sample assists the target field to perform word segmentation, so that the sample labeling time is shortened, and the word segmentation performance can be improved at the same time.

Description

technical field [0001] The invention relates to the technical field of joint recommendation systems, in particular to a joint recommendation system based on a joint learning framework. Background technique [0002] Chinese NER entity recognition requires a large amount of corpus for model training, but in reality, the corpus data is lacking, requiring a lot of manpower and material resources for data annotation. [0003] Patent No. CN202010895913.X discloses a Chinese word segmentation and entity recognition joint learning method automatically generated by a data set, the method includes the following steps: the first step, the construction of the target domain data set; the second step, the first step The sentence s with a string of Chinese character sequences in the data set obtained in is input into the character vector representation layer of the neural network model, and the vector representation of each Chinese character is obtained; the third step, the Chinese charact...

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): G06N20/00G06N3/04G06N3/08G06F40/295
CPCG06N20/00G06N3/049G06N3/08G06F40/295
Inventor 赵蕾
Owner ENNEW DIGITAL 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