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.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
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...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


