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

Accelerated k-means clustering

A technology for input data and sample data sets, applied in complex mathematical operations, character and pattern recognition, instruments, etc., can solve problems such as K-means clustering calculation efficiency needs to be improved

Inactive Publication Date: 2018-12-21
HUAWEI TECH CO LTD
View PDF0 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0035] Despite the continuous improvement of existing technology, the computational efficiency of K-means clustering still needs to be improved

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
  • Accelerated k-means clustering
  • Accelerated k-means clustering
  • Accelerated k-means clustering

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0086] figure 1 An apparatus 100 for iterative clustering of an input data set is shown.

[0087] The device 100 for iteratively clustering an input data set includes a direction updating unit 110 , a step updating unit 120 and a center updating unit 130 . The units 110 to 130 can be implemented on a processor such as ( figure 1 not shown).

[0088] The direction updating unit 110 is configured to calculate an updated direction of the current iteration according to one or more cluster centroids of one or more previous iterations.

[0089] The cluster centroids of previous iterations can be defined as in is the element of the previous iteration's cluster (i is the index of the current iteration). Therefore, the direction update unit 110 may calculate an update direction based on the cluster members of previous iterations, that is, the direction from the previous cluster center to the current cluster center.

[0090] The step size update unit 120 is used for determining ...

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

A device for iteratively clustering an input dataset, the device comprising a direction update unit configured to compute an update direction of a current iteration based on one or more cluster centroids of one or more previous iterations, a step size update unit configured to de termine an update step size factor based on a Tchebychev polynomial, and a center update unit configured to compute anupdated cluster center based on the update direction and the update step size factor.

Description

technical field [0001] The invention relates to a device for iterative clustering of an input data set and an iterative method for clustering an input data set. The present invention also relates to a computer-readable storage medium storing program code, the program code including instructions for executing the above method. Background technique [0002] K-means clustering techniques are widely used in computer vision, machine learning, genomics and bioinformatics, medical cybernetics, and financial applications, such as: [0003] 1) for direct clustering of big data; [0004] 2) Used in computer graphics for color quantization (reducing the palette of an image to a fixed number of k colors); [0005] 3) as an initial program to accelerate other ML techniques; [0006] 4) as a preliminary procedure for accelerating clustering methods for community detection; [0007] 5) As a preliminary program for the parallelized support vector clustering method. [0008] Especially ...

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
IPC IPC(8): G06F17/10G06K9/62
CPCG06F17/10G06F18/23213G06F18/24137
Inventor 米哈伊尔·彼得罗维奇·莱文
Owner HUAWEI TECH CO LTD