# US 2025/0156501 A1 · Chart pattern recognition system and method

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2025/0156501 A1
- Document ID: US20250156501A1
- Status: Published application
- Kind: A1
- Application number: 18/733,892
- Filed: 2024-06-05
- Published: 2025-05-15
- Earliest claimed: 2023-11-10
- Assignees: National Tsing Hua University
- Applicants: National Tsing Hua University
- Inventors: Chuan-Hsiang Han; Jyh-Shing Jang
- Web page: https://patentgrep.com/patent/US20250156501A1

## CPC

- G06F 18/22: Matching criteria, e.g. proximity measures
- G06F 18/10: Pre-processing; Data cleansing
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

A chart pattern recognition method includes pre-processing a chart pattern group to be compared to obtain the first chart pattern data to be compared based on time and space; normalizing the first chart pattern data to be compared and the first historical chart pattern data from a database to obtain the second chart pattern data to be compared and the second historical chart pattern data; comparing the second chart pattern data to be compared with the second historical chart pattern data to obtain multiple similarity scores of the second chart pattern data to be compared and the second historical chart pattern data; sorting the multiple similarity scores, the corresponding second chart pattern data to be compared, and the corresponding first historical chart pattern data; and obtaining first historical chart pattern data with a maximum similarity score.

## Claims

### Claim 1

A chart pattern recognition system, comprising:
a database, configured to store multiple pieces of first historical chart pattern data;
a non-volatile memory, configured to store a chart pattern recognition method; and
a processing circuit, coupled to the database and the non-volatile memory and configured to execute the chart pattern recognition method, wherein the chart pattern recognition method comprises steps of:
pre-processing a chart pattern group to be compared to obtain multiple pieces of first chart pattern data to be compared based on time and space;
normalizing the multiple pieces of first chart pattern data to be compared and the multiple pieces of first historical chart pattern data to obtain multiple pieces of second chart pattern data to be compared and multiple pieces of second historical chart pattern data;
comparing the multiple pieces of second chart pattern data to be compared with the multiple pieces of second historical chart pattern data to obtain multiple similarity scores between the multiple pieces of second chart pattern data to be compared and the multiple pieces of second historical chart pattern data;
sorting the multiple similarity scores with the corresponding multiple pieces of second chart pattern data to be compared and the corresponding multiple pieces of first historical chart pattern data; and
obtaining first historical chart pattern data with a maximum similarity score.

### Claim 2 (depends on 1)

The chart pattern recognition system according to claim 1, wherein the step of comparing the multiple pieces of second chart pattern data to be compared with the multiple pieces of second historical chart pattern data comprises:
transforming the multiple pieces of second chart pattern data to be compared and the multiple pieces of second historical chart pattern data through Fourier transform to compare the multiple pieces of second chart pattern data to be compared with the multiple pieces of second historical chart pattern data to obtain the multiple similarity scores.

### Claim 3 (depends on 1)

The chart pattern recognition system according to claim 1, wherein the step of comparing the multiple pieces of second chart pattern data to be compared with the multiple pieces of second historical chart pattern data comprises:
comparing the multiple pieces of second chart pattern data to be compared with the multiple pieces of second historical chart pattern data on a normalized vector through a query by singing/humming method to obtain the multiple similarity scores.

### Claim 4 (depends on 1)

The chart pattern recognition system according to claim 1, wherein the step of pre-processing the chart pattern group to be compared to obtain the multiple pieces of first chart pattern data to be compared based on a time period comprises:
vectorizing the chart pattern group to be compared to obtain third chart pattern data to be compared based on time; and
linearly scaling the third chart pattern data to be compared based on the time and space to obtain the multiple pieces of first chart pattern data to be compared.

### Claim 5 (depends on 1)

The chart pattern recognition system according to claim 1, wherein the chart pattern group to be compared comprises a compared line and a reference line.

### Claim 6

A chart pattern recognition method stored in a non-volatile memory and executed by a processing circuit, the chart pattern recognition method comprising steps of:
pre-processing a chart pattern group to be compared to obtain multiple pieces of first chart pattern data to be compared based on time and space;
normalizing the multiple pieces of first chart pattern data to be compared and multiple pieces of first historical chart pattern data from a database to obtain multiple pieces of second chart pattern data to be compared and multiple pieces of second historical chart pattern data;
comparing the multiple pieces of second chart pattern data to be compared with the multiple pieces of second historical chart pattern data to obtain multiple similarity scores of the multiple pieces of second chart pattern data to be compared and the multiple pieces of second historical chart pattern data;
sorting the multiple similarity scores, the multiple pieces of second chart pattern data to be compared corresponding to the multiple similarity scores, and the multiple pieces of first historical chart pattern data corresponding to the multiple pieces of second historical chart pattern data corresponding to the multiple pieces of second chart pattern data to be compared; and
obtaining first historical chart pattern data with a maximum similarity score.

### Claim 7 (depends on 6)

The chart pattern recognition method according to claim 6, wherein the step of comparing the multiple pieces of second chart pattern data to be compared with the multiple pieces of second historical chart pattern data comprises:
transforming the multiple pieces of second chart pattern data to be compared and the multiple pieces of second historical chart pattern data through Fourier transform to compare the multiple pieces of second chart pattern data to be compared with the multiple pieces of second historical chart pattern data to obtain the multiple similarity scores.

### Claim 8 (depends on 6)

The chart pattern recognition method according to claim 6, wherein the step of comparing the multiple pieces of second chart pattern data to be compared with the multiple pieces of second historical chart pattern data comprises:
comparing the multiple pieces of second chart pattern data to be compared with the multiple pieces of second historical chart pattern data on a normalized vector through a query by singing/humming method to obtain the multiple similarity scores.

### Claim 9 (depends on 6)

The chart pattern recognition method according to claim 6, wherein the step of pre-processing the chart pattern group to be compared to obtain the multiple pieces of first chart pattern data to be compared based on a time period comprises:
vectorizing the chart pattern group to be compared to obtain third chart pattern data to be compared based on time; and
linearly scaling the third chart pattern data to be compared based on the time and space to obtain the multiple pieces of first chart pattern data to be compared.

### Claim 10 (depends on 6)

The chart pattern recognition method according to claim 6, wherein the chart pattern group to be compared comprises a compared line and a reference line.
