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

A method based on multi-model stack fusion prediction

A forecasting method and multi-model technology, applied in the field of machine learning, can solve problems such as the inability to publicly obtain the real financial information of enterprises, and achieve the effect of good business exit risks and improved stability

Inactive Publication Date: 2019-03-26
SUN YAT SEN UNIV
View PDF0 Cites 9 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, for a larger number of small and medium-sized enterprises, neither the real financial information of the enterprises nor the public credit information of these enterprises can be obtained publicly. What can be collected may only be the behavioral footprint information data left by small and medium-sized enterprises in various aspects. The business risk prediction of small and medium-sized enterprises has brought great challenges

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
  • A method based on multi-model stack fusion prediction

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0025] S1: Select 5 single models from 7 single models as the first-level prediction model;

[0026] S2: Divide the data sample into five parts, take four of them as the training set and one as the test set, respectively, make predictions on the five single models, and the test set of each single model is different;

[0027] S3: The prediction result of the j-th single model on the i-th training sample will be used as the j-th eigenvalue of the i-th sample in the new training set, and as the new training set, where

[0028] S4: The prediction result of the j-th single model on the i-th test sample will be used as the j-th eigenvalue of the i-th sample in the new test set, and all test results will be input to the two-layer single model as the new test set. Test where

[0029] S5: Set three different random seeds for each single model to test the same model three times, take the arithmetic average of the three predicted results as the predicted value of the model, and use the predict...

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 based on multi-model stacking fusion prediction, which aims at solving the problem of using weak variable to objectively and fairly evaluate the operation status of anenterprise in the absence of strong variable, and predicting the operation risk status of the enterprise in a period of time in the future. At first, that invention carry out data analysis on the enterprise operation behavior data, and carries out data preprocess and feature extraction on the enterprise operation behavior data; Using the extracted features to train several single models and verify the effectiveness of the model, select the optimal single model for stack fusion; The probability value of enterprise operation risk is predicted by the method of multi-single-model stacking fusion.This method has a good ability to predict the business exit risk of enterprises.

Description

Technical field [0001] The present invention relates to the field of machine learning, and more specifically, to a method based on multi-model stacking fusion prediction. Background technique [0002] Traditional enterprise evaluation is mainly based on the financial information of the enterprise, loan record information, etc. to judge the business status of the enterprise, and the credit information such as the possibility of default. For large and medium-sized enterprises with sound finances and records in the traditional bank lending field, this evaluation method is undoubtedly more objective and reasonable. However, for a larger number of small, medium and micro enterprises, neither the real financial information of the enterprises nor the public credit information of these enterprises can be publicly obtained. What can be collected may only be the behavioral footprint information data left by the small and medium enterprises in various aspects. The business risk prediction ...

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): G06K9/62G06Q10/06
CPCG06Q10/0635G06F18/24323G06F18/214
Inventor 郑子彬曾璇
Owner SUN YAT SEN UNIV
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