Method for building LS-SVM prediction model based on chaotic search

A chaotic search and predictive model technology, applied in database models, structured data retrieval, special data processing applications, etc., can solve the problems of network infrastructure network resources such as endless demand, excessive use of resources, idle resources, etc.

Inactive Publication Date: 2014-12-10
GUILIN UNIV OF ELECTRONIC TECH
View PDF0 Cites 8 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0002] The improvement of network infrastructure cannot meet people's endless demand for network resources, which leads to many undesirable phenomena such as network per

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
  • Method for building LS-SVM prediction model based on chaotic search
  • Method for building LS-SVM prediction model based on chaotic search
  • Method for building LS-SVM prediction model based on chaotic search

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0059] The present invention will be described in more detail below in conjunction with the accompanying drawings.

[0060] see figure 1 , the model structure of the present invention includes five parts: establishment of sample set, calculation of model coefficients, optimization of model parameters, determination of prediction model and sample update processing.

[0061] The method of the present invention first establishes a sample set to realize data sampling of the prediction object, and establishes a sample training data set. According to the operation status of the prediction object, the operation status of the prediction object such as computer server is sampled according to a certain frequency (such as sampling every 5 minutes), including CPU utilization rate, memory usage rate, etc. For the collected data, the following data processing is performed: ① Eliminate extreme values, that is, remove values ​​that deviate far from the sample; ② Determine the length of the t...

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

The invention relates to a method for building an LS-SVM prediction model based on the chaotic search. The method includes the following steps: (A) building a sample training data set; (B) calculating coefficients of the model; (C) conducting optimization with the chaotic search improvement algorithm, and obtaining the minimum value and the optimal chaotic variable of a to-be-optimized function; (D) determining the optimized LS-SVM prediction model; (E) updating a sample. By means of the method, the LS-SVM self-adaptation resource prediction model is built after parameters of the model are optimized with the chaotic search improvement algorithm, the operating state of prediction objects in cloud calculation can be dynamically predicted, prediction results have good adaptability, and it can be guaranteed that the prediction resultsmoreapproximate to true values of the prediction objects. The sensibility of the chaotic search to the initial value is remitted through the model; in addition, in the chaotic iterative search process, the second-time search can be rapidly carried out in the optimal solution neighborhood through the adjustment on the chaotic variable, the search efficiency is improved, and the possibility of being caught into the local optimum is decreased.

Description

technical field [0001] The invention relates to a resource prediction technology, in particular to a method for establishing a chaotic search-based LS-SVM prediction model. Background technique [0002] The improvement of network infrastructure cannot meet people's endless demand for network resources, which leads to many undesirable phenomena such as network performance degradation and response delay. In fact, in the network, on the one hand, some resources are overused, and on the other hand, a large number of resources are idle. Cloud computing uses virtualization technology to construct resource pools of various underlying physical resources in the network, and distributes user requests to different resource pools for on-demand services and on-demand billing. How to reasonably allocate and schedule underlying physical resources in the cloud platform, effectively improve service quality and system performance, and ensure load balancing has become a challenge for the indu...

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): G06F17/30
CPCG06F16/2468G06F16/28
Inventor 张润莲武小年张明玲李园园杨宇洋
Owner GUILIN UNIV OF ELECTRONIC TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap