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

Stock selection method based on relation-time sequence diagram convolution

A timing diagram and relationship technology, which is applied in neural learning methods, instruments, biological neural network models, etc., can solve the problem of not being able to consider stock relationship and timing characteristics at the same time, and achieve the effect of gradient optimization and fast training speed

Pending Publication Date: 2021-06-11
四川省人工智能研究院(宜宾)
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] Aiming at the above-mentioned deficiencies in the prior art, the present invention provides a stock selection method based on relational-sequence graph convolution, and the present invention solves the problem that the relational-sequential features of stocks cannot be considered at the same time in stock forecasting in the past

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
  • Stock selection method based on relation-time sequence diagram convolution
  • Stock selection method based on relation-time sequence diagram convolution
  • Stock selection method based on relation-time sequence diagram convolution

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0050] Such as figure 1 As shown, the present invention provides a stock selection method based on relational-sequence graph convolution, and its implementation method is as follows:

[0051] S1. Using the timing characteristics of stocks and the extracted external relationships, construct a relationship-sequence diagram based on the entire stock market, specifically: according to the timing characteristics of stocks and the extracted external relationships, hierarchical the timing characteristics and external relationships of several stocks Representation, build a relationship-sequence diagram based on the entire stock market, the relationship-sequence diagram is composed of T relationship diagrams, each of the relationship diagrams has N nodes, and each node represents a stock; and each of the relationships- The timing diagram includes relationship edges and time edges; relationship edges are used to represent the relationship between stocks; timing edges are used to connect...

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 provides a stock selection method based on relation-time sequence diagram convolution, which belongs to the technical field of stock selection and comprises the following steps: constructing a relation-time sequence diagram based on a whole stock market; based on the relation-time sequence diagram of the whole stock market, extracting relation-time sequence features of each stock by using a relation-time sequence diagram convolutional network in combination with a pooling layer; according to the extracted relation-time sequence features of each stock, calculating a predicted return rate of each stock by using a full connection layer, and optimizing a relation-time sequence diagram convolutional network; and based on the optimized relation-time sequence diagram convolutional network, sorting all stock prediction return rates in the stock market from high to low, and selecting the first N stocks with the highest return rate. The stock trend prediction not only needs to consider the time sequence information of each stock, but also needs to consider other stock information associated with the stock in the market, so that the problem that the relationship-time sequence features of stocks are not considered at the same time during stock trend prediction in the prior art is solved through the design.

Description

technical field [0001] The invention belongs to the technical field of stock selection, and in particular relates to a stock selection method based on relation-sequence graph convolution. Background technique [0002] Stock forecasting is a research hotspot in the field of financial technology, and whether an investor can make profits in the stock market depends largely on whether he / she can buy or trade stocks at the right time. Therefore, accurate forecasting of stock movements is the key to making correct investment decisions. With the help of computer technology, many algorithms have been used to predict stock movements. For example, the paper "Liheng Zhang, Charu C. Aggarwal, and Guo-Jun Qi, "Stock Price Prediction via Discovering Multi-Frequency Trading Patterns," In Proceedings of the 23rdACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Halifax, NS, Canada ,August 13-17,2017, pp.2141–2149" proposed to use State Frequency Memory (SFM) to ca...

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): G06Q40/04G06Q10/04G06N3/08G06N3/04
CPCG06Q40/04G06Q10/04G06N3/08G06N3/045
Inventor 邵杰郑泽涛申恒涛
Owner 四川省人工智能研究院(宜宾)
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