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

Transaction stock rapid finding method based on time sequence convolution self-coding

A convolutional self-encoding and timing technology, applied in neural learning methods, data processing applications, finance, etc., can solve problems such as losses, investors missing trading opportunities, and reduced profits

Pending Publication Date: 2019-10-18
SHANDONG UNIV OF FINANCE & ECONOMICS
View PDF0 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

If you search for abnormal stocks manually or only by index thresholds, you can often only find out at a later stage, and investors will miss trading opportunities, resulting in reduced profits or even losses

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
  • Transaction stock rapid finding method based on time sequence convolution self-coding
  • Transaction stock rapid finding method based on time sequence convolution self-coding
  • Transaction stock rapid finding method based on time sequence convolution self-coding

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0076] The present invention provides a method for quickly finding abnormal stocks based on time-series convolutional self-encoding, including:

[0077] S1. Obtain stock trading market data within a time span; stock trading market data such as a trading day, an economically related securities market - such as China's Shanghai and Shenzhen stock markets;

[0078] S2. Formalize the time-series data of the stock trading market data obtained in S1 to generate time-series formalized data of the stock;

[0079] S3. Generate a time-series convolutional self-encoding model according to the time-series formalized data of S2, and train the time-series convolutional self-encoding model;

[0080] S4. According to the time-series convolutional autoencoder model obtained in S3, combined with the time-series formalized data of S2, find abnormal stocks. Abnormal stocks refer to a small number of stocks that are significantly different from the trend of most stocks.

[0081] In S2, the forma...

Embodiment 2

[0111] see figure 1 , on the basis of Embodiment 1, a system for quickly finding abnormal stocks based on time-series convolutional self-encoding is proposed, including a transaction data acquisition module, a time-series data formal representation module, a model generation module, and an abnormal stock detection module;

[0112] The data input terminal of the transaction data acquisition module is connected to the stock database, and the transaction data acquisition module, the time series data formal representation module, and the transaction stock detection module are connected in sequence to form a data link;

[0113] Wherein the data output end of the transaction data acquisition module is connected to the data input end of the time series data formal representation module; the data output end of the time series data formal representation module is respectively connected to the input ends of the mobile stock detection module and the model generation module;

[0114] The ...

Embodiment 3

[0116] On the basis of Embodiment 2, a method for quickly finding abnormal stocks based on the time-series convolutional self-encoding system for quickly finding abnormal stocks is proposed,

[0117] Step 1, the transaction data acquisition module is responsible for obtaining all stock transaction data within a time span - such as a trading day, an economically related securities market - such as China's Shanghai and Shenzhen stock markets; after obtaining the data, it is imported in the form of time series data representation module;

[0118] Step 2, the time series data formal representation module is responsible for the formal representation of the incoming transaction data;

[0119] Step 3, the model generation module generates a time-series convolutional self-encoding model according to the time-series formalized data in step 2, and trains the time-series convolutional self-encoding model;

[0120] Step 4: The abnormal stock detection module finds stocks with abnormal tr...

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 transaction stock rapid finding method based on time sequence convolution self-coding, and relates to the technical field of data mining, and the technical scheme is that themethod comprises the steps: obtaining stock transaction quotation data in a time span; performing time series data formalized representation on the obtained stock transaction quotation data to generate time series formalized data of the stock; generating a time sequence convolution self-encoding model according to the time sequence formalized data, and training the time sequence convolution self-encoding model; and finding out the transaction stock according to the time sequence convolution self-encoding model in combination with the time sequence formalized data. The beneficial effects of the invention are that the method can automatically detect the transaction stock in any time span, and saves the manual analysis cost; 2, transaction stock s with non-breakthrough index threshold typescan be detected.

Description

technical field [0001] The present invention relates to the technical field of data mining, in particular to a system and method for quickly finding abnormal stocks based on time-series convolutional self-encoding. Background technique [0002] Abnormal stocks refer to a small group of stocks that have different trading characteristics from the majority of stocks in a certain time span. In the same stock exchange market or the same economy, the trends of most stock prices, trading volume and other indicators are basically similar. However, due to the influence of news or the intentional actions of major trading participants, a small number of abnormal stocks will have independent trends. In a short period of time, there is a high probability that the price of abnormal stocks will fluctuate significantly. Finding abnormal stocks among a large number of stocks in the market, together with forecasting models and financial engineering tools, will help investors reduce losses a...

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): G06Q40/04G06N3/08
CPCG06Q40/04G06N3/084
Inventor 邹立达耿蕾蕾蔡慧英冉令强马艳
Owner SHANDONG UNIV OF FINANCE & ECONOMICS
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