Apparent density peak detection algorithm based on double-check

Abstract

The invention relates to an apparent density peak detection algorithm based on double-check. The apparent density peak detection algorithm based on double-check comprises the following steps that: a local density peak point is found out by calculating the distance between every two data concentration points to be clustered and outer densities of all points; and then, an apparent density peak point is finally found out by calculating the inner density of the local density peak point. Because of calculation of the inner density, the flexibility and the complex data processing capability of a density-based clustering algorithm are improved; the apparent density peak detection algorithm based on double-check can effectively increase the accuracy rate of the traditional density-based clustering algorithm and improve the bit error rate performance of the system; and thus, a relatively low error floor can be obtained.

Description

technical field [0001] The invention relates to the field of computer operation, in particular to a false density peak detection algorithm based on double inspection. Background technique [0002] Clustering is a process of classifying and distinguishing physical or abstract objects according to their similarities. For decades, clustering has been one of the important research contents in data mining, pattern recognition and other research directions. Traditional density-based clustering algorithms, such as DBSCAN, usually have better clustering results. However, due to the fixed scanning radius and insufficient flexibility of this type of algorithm, the density peak points found by this type of algorithm may have flaws and false peak points, that is, some points will be mistaken for peak points, so when dealing with some complex When the shape and distribution of data are disordered, the error rate of classification will increase. Therefore, dealing with these false peak ...

AI Technical Summary

Problems solved by technology

However, due to the fixed scanning radius and insufficient flexibility of this type of algorithm, the density peak points found by this type of algorithm may have flaws and false peak points, that is, some points will be mistaken for peak points, so when dealing with some complex When the shape and distribution of data are disordered, the error rate of classification will increase. Therefore, dealing with these false peak points will help improve the accuracy of clustering results. Traditional density-based clustering algorithms, such as DBSCAN, etc., due to Its scanning radius is fixed, and the result is not good when dealing with complex structured data

Images