Method and device for constructing model based on PU learning and prediction method and device

A model and target model technology, applied in the field of machine learning, can solve the problems of difficult models and high thresholds, and achieve the effect of lowering thresholds, reducing labor costs, and reducing human consumption

Pending Publication Date: 2019-08-02
THE FOURTH PARADIGM BEIJING TECH CO LTD
View PDF7 Cites 16 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] In view of the above problems, the present invention proposes a method and device for building a model based on PU learning, the main purpose of which is to solve the difficulty of model construction caused by the need for operators to have certain business experience in the existing PU learning model training process Larger, higher-threshold problems, thereby lowering the threshold for building models

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
  • Method and device for constructing model based on PU learning and prediction method and device
  • Method and device for constructing model based on PU learning and prediction method and device
  • Method and device for constructing model based on PU learning and prediction method and device

Examples

Experimental program
Comparison scheme
Effect test

example 1

[0088] The present invention also provides a method for implementing target object recommendation, the method comprising:

[0089] 11) According to figure 1 The method obtains the target model, wherein the sample data set for training the target model is a data set about the target object, wherein the data of the target object selected by the user is positive sample data, and the data of the target object not selected by the user The data of the object is unlabeled sample data;

[0090] 12) Acquiring target object data to be predicted, wherein the target object is a commodity or service provided through the Internet.

[0091] 13) Use the obtained target model to execute the target object recommendation task, specifically, use the target model to predict the recommended value (such as the recommendation probability) of the target object to be predicted.

[0092] In the process of realizing the target object recommendation, the target object data to be predicted may be obtaine...

example 2

[0120] The present invention also provides a method for detecting bank card leakage points, the method comprising:

[0121] 21) According to figure 1 The method obtains a target model, wherein the sample data set for training the target model is a transaction data set of a stolen bank card, wherein the transaction data of a stolen bank card whose leakage point has been marked is positive sample data , the transaction data of the stolen bank card that has not marked the leakage point is the unmarked sample data;

[0122] 22) Obtain the transaction data of the stolen bank card at the leakage point to be detected.

[0123] 23) Use the obtained target model to perform the task of bank card leakage point detection.

[0124] In the process of training the model to realize the detection of bank card leakage points, the historical transaction data of stolen bank cards can be obtained first.

[0125] Second, in the historical data, the transaction data of the known leakage point is ...

example 3

[0138] The present invention also provides a method for realizing image / text classification, the method comprising:

[0139] 31) According to figure 1 The method obtains the target model, wherein the sample data set for training the target model is an image / text data set, wherein the image / text data with classification marks are positive sample data, and the image / text data without classification marks is unlabeled sample data; the target model is used to perform image / text classification tasks;

[0140] 32) Obtain image / text data to be predicted.

[0141] 33) Use the obtained target model to perform image / text classification tasks.

[0142] In the process of realizing image / text classification, it may specifically be as follows: First, acquire image / text data to be predicted.

[0143] Second, obtain image / text historical data, which includes classified data as positive sample data and unclassified data as unlabeled sample data, and obtains this historical data as sample da...

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 method and device for constructing a model based on PU learning, and a prediction method and device, relates to the technical field of machine learning, and mainly aims to solve the problems of high model construction difficulty and high threshold caused by the fact that an operator needs to have a certain service experience in the existing PU learning model training process. The main technical scheme of the invention is as follows: obtaining a sample data set, wherein the sample data set comprises positive sample data with a positive label and unlabeled sample data without a label; performing training based on the sample data set to obtain a plurality of candidate models; constructing an evaluation set based on the sample data set; evaluating each candidate modelaccording to the evaluation set and a preset evaluation condition to obtain an evaluation result corresponding to each candidate model; selecting a candidate model with an evaluation result meeting apreset condition; and integrating the selected model according to a preset integration method to obtain a target model. The method is used for constructing the PU learning model.

Description

technical field [0001] The present invention relates to the technical field of machine learning, in particular to a method and device for constructing a model based on PU learning, and a method and device for predicting using the model. Background technique [0002] With the continuous advancement of technology, artificial intelligence technology is also gradually developing. Among them, machine learning is an inevitable product of the development of artificial intelligence research to a certain stage. It is committed to improving the performance of the system itself by means of computing and using experience. In computer systems, "experience" usually exists in the form of "data". Through machine learning algorithms, "models" can be generated from data. Model, when faced with a new situation, the model will provide the corresponding judgment, that is, predict the result. Whether training a machine learning model or making predictions using a trained machine learning model,...

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 涂威威王海
Owner THE FOURTH PARADIGM BEIJING 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