Complex network construction method based on combination of mapping and density peak clustering

Abstract

The embodiment of the invention provides a complex network construction method based on combination of mapping and density peak clustering. The method comprises the steps of preprocessing a time sequence through a preprocessing algorithm combining mapping and density peak clustering; Constructing complex networks. According to the method, a preprocessing algorithm combining mapping and density peak clustering is used for optimizing construction of a complex network, the fluctuation mode of a time sequence can be described more accurately, effective information in data can be mined more deeply,and higher robustness is achieved on noise.

Description

technical field [0001] This application relates to the technical field of complex network construction, in particular to a complex network construction method based on the combination of mapping and density peak clustering. Background technique [0002] As a method widely used in recent years, complex network has made achievements in many fields, especially in the mining of time series features, which has wide applicability. More importantly, complex networks can accurately describe the fluctuation patterns of time segments while describing the overall fluctuations, which helps to understand the fluctuation patterns and laws of time series more clearly. However, since the actual data often contains a lot of noise, directly using the original time series to construct the network will greatly interfere with the analysis results, so it is necessary to preprocess the data so that the nodes can express the key fluctuations in the time series more accurately and effectively model...

AI Technical Summary

Problems solved by technology

This method can remove interference to a certain extent and indicate the size characteristics of the time series in a symbolic way, but it lacks the description of the transformation between sequence modes, and requires the data to obey the Gaussian distribution, which is not universal

In addition, the PAA algorithm smoothes out some key peak points while removing noise, which will change the overall fluctuation pattern and adversely affect the analysis results.

Qualitative Trend Analysis (QTA) is used in the description of time series. This method accurately represents the fluctuation pattern of time series in a symbolic way, but it cannot rule out the interference of redundant fluctuations, which will produce many unnecessary fluctuations. model

Images