Method for adaptively adjusting clustering center label in swarm intelligence algorithm
A self-adaptive adjustment and clustering center technology, applied in text database clustering/classification, calculation, calculation model, etc., can solve problems such as unsatisfactory algorithm update efficiency, improve population update efficiency and global search ability, and improve population Effect of Update Efficiency and Convergence Speed
Pending Publication Date: 2019-10-18
GUANGDONG UNIV OF TECH
View PDF0 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0004] Aiming at the problem of unsatisfactory algorithm update efficiency caused by the inconsistency of the cluster center labels of population individuals in the process of solving the optimal clustering center of the clustering problem in the swarm intelligence evolutionary algorithm, the present invention provides an adaptive adjustment in the swarm intelligence algorithm A method of clustering center labeling to improve the update efficiency and global search ability of the swarm intelligence evolutionary algorithm in the process of text clustering
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment 1
[0035] This embodiment provides a method for adaptively adjusting the label of the cluster center in the swarm intelligence algorithm, such as figure 1 , including the following steps:
[0036] S1: Set an individual x based on its cluster center label j , another individual whose cluster center label needs to be adjusted is x i , to calculate the individual x j with individual x i The similarity matrix M of each cluster center in
[0037] S2: According to the principle of maximum similarity, the index list of the maximum value of each row of the statistical similarity matrix M and the index list of the maximum value of each column represent the individual x i expectations and individual x j The cluster centers of are expected to form a list L of correspondences ij , individual x j expectations and individual x i The cluster centers of form a list L of corresponding relations ji ;
[0038] S3: Judgment L ij Whether there are duplicate values, that is, to judge the in...
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
Login to View More Abstract
The invention discloses a method for adaptively adjusting a clustering center label based on two groups of clustering center similarity matrixes, for solving the problem that when a swarm intelligenceevolutionary algorithm adopts a division-based form for clustering, clustering center labels are inconsistent due to random arrangement of clustering centers among individuals in a constructed population, that is, the clustering centers belonging to different clusters are arranged in the same dimension, so that the problems of low algorithm population individual updating efficiency and search blindness are caused. According to the method, a bidirectional selection competition elimination strategy is adopted, and clustering centers with the maximum similarity and the closest distance between individuals are arranged in the same dimension as much as possible, and the maximization of clustering center label consistency is guaranteed.
Description
technical field [0001] The invention relates to the fields of clustering algorithms and swarm intelligence evolution algorithms, and more specifically, relates to a method for adaptively adjusting cluster center labels in swarm intelligence algorithms. Background technique [0002] Text clustering plays a vital role in the field of text mining. It has always been one of the common needs of enterprises or government agencies to effectively organize and divide massive text data in the Internet and actual production environments, and to discover hidden valuable information. . Text is the most natural way to store information. It is a special kind of unstructured data with the characteristics of high dimensionality, sparse features and low data correlation. Currently commonly used clustering algorithms, such as K-means, K-means++, etc., are highly sensitive to the selection of the initial cluster center, and the way of adjusting the cluster center also makes the algorithm easy ...
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
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G06F16/35G06K9/62G06N3/00
CPCG06F16/35G06N3/006G06F18/23211
Inventor 胡晓敏王明丰李瑞珠李敏罗玉
Owner GUANGDONG UNIV OF TECH