Unlock instant, AI-driven research and patent intelligence for your innovation.

A Pattern Extraction and Evolution Visual Analysis Method Based on Time Series Multivariate Data

An analysis method and multivariate technology, applied in the field of data visualization and visual analysis, which can solve the problems of difficult comparison of results, low calculation efficiency, and reduced analysis accuracy.

Active Publication Date: 2022-07-15
NORTHEAST NORMAL UNIVERSITY
View PDF14 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The above method is to reduce the dimensionality of data in different time slices to the same space, but due to the large data size, the calculation efficiency is low, and the redundant information between different time slices will reduce the analysis accuracy
Another approach is to reduce the dimensionality of the data on different time slices separately, but it will cause the data points of different time slices to belong to different spaces, and the results are difficult to compare
Therefore, there is currently a lack of a dimensionality reduction method that can handle time-varying features and lay the foundation for analyzing patterns and anomalies in time-series multivariate data.

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 Pattern Extraction and Evolution Visual Analysis Method Based on Time Series Multivariate Data
  • A Pattern Extraction and Evolution Visual Analysis Method Based on Time Series Multivariate Data
  • A Pattern Extraction and Evolution Visual Analysis Method Based on Time Series Multivariate Data

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0041] Example 1: Please refer to figure 1 ,

[0042] Regular mode: A set of data whose category remains unchanged in all time slices is called regular mode, wherein different categories determine different regular modes.

[0043]But for outliers in time series multivariate data, there is no valid definition so far. Based on some existing anomaly detection work, we comprehensively consider the time-varying patterns of various data in terms of anomaly pattern exploration, starting from the opposite of conventional patterns. According to the fluctuation of anomalies in time series, they are divided into stable anomalies and jumping anomalies; according to the changes in the neighborhood, they are divided into isolated anomalies and cooperative anomalies. Combining the abnormal characteristics of data in time series and neighborhood, we extract four abnormal patterns; stable isolated anomaly: the sample point that always deviates from most other data in all time slices is calle...

Embodiment 2

[0070] The system is developed on a personal computer, the specific environment is windows10 64-bit operating system, 16GB running memory, AMD Ryzen 7 4800H with Radeon Graphics 2.90GHz processor, GeForce GTX 1650 graphics card. The data analysis part of the system is written in Python language, and the front-end visualization interface uses D3.js data visualization graphics library, Echarts interactive chart and browser visualization library. During the experimental evaluation, we use a Samsung monitor with a resolution of 1920 × 1080 and use the Chrome browser as the front-end to display the application.

[0071] The data set used is the national consumer price index (CPI) data, which is downloaded from the website of the National Bureau of Statistics (https: / / data.stats.gov.cn / ). Changes in time, changes in prices, and relative numbers that reflect changes in the price levels of consumer goods and services purchased by residents (the base period value of the index is set at...

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 present invention relates to the field of data visualization and visual analysis, and aims to provide a pattern extraction and evolution visual analysis method based on time series multivariate data, comprising the following steps: S1: Process the data sample points to be processed, and establish a time series multivariate data sample point. For the variable data set, normal points and abnormal points are filtered out by anomaly detection algorithm, and step 2 is performed; S2: Extract normal points by normal mode, and select abnormal points by abnormal mode, and perform step 3; S3: Use visual analysis system PEVis establishes multiple interconnected views and displays them through front-end interaction. On the basis of maintaining the time consistency of patterns and anomalies in the data, a novel visual representation scheme is designed to help users intuitively perceive the evolution of regular patterns and abnormal events.

Description

technical field [0001] The invention relates to the technical field of data visualization and visual analysis, in particular to a pattern extraction and evolution visual analysis method based on time series multivariate data. Background technique [0002] With the continuous improvement of urban informatization and the advent of the era of big data, the data generated by social development and human life have been widely recorded and collected. Multivariate data is a very common type of data, and its data samples have multiple attribute characteristics, such as environmental monitoring data containing multiple indicators, personal files containing multiple information, etc. Time is constantly changing, which is called time series multivariate data, and analyzing the underlying patterns contained in time series multivariate data can reflect the changing laws of things in the real world to a large extent. [0003] In addition, in the fields of anti-cheating, fake base station...

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 Patents(China)
IPC IPC(8): G06F17/16G06K9/62
CPCG06F17/16G06F18/23213
Inventor 张慧杰吕程任珂付佳蔺依铭
Owner NORTHEAST NORMAL UNIVERSITY