Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
EXTREME LEARNING MACHINE METHOD FOR IMPROVING ARTIFICIAL BEE COLONY OPTIMIZATION
Document Type and Number:
WIPO Patent Application WO/2017/197626
Kind Code:
A1
Abstract:
An extreme learning machine method for improving artificial bee colony optimization. The method comprises the following steps: step 1, generating initial solutions for SN entities; step 2, performing global optimization with respect to a connection weight ɷ and a threshold b of the extreme learning machine; step 3, performing local optimization with respect to the connection weight ɷ and the threshold b of the extreme learning machine; step 4, if food source information is not updated within a fixed duration, converting an employee bee into an investigation bee and returning to step 1 to re-initialize the entity; and step 5, extracting the connection weight ɷ and the threshold b of the extreme learning machine from an optimal entity and using a test set to perform verification. The method can be used to better overcome poor results experienced when traditional extreme learning machines are applied to classification and regression. In comparison to traditional extreme learning machines and the SaE-ELM algorithm, the method is more robust and can effectively improve the results of classification and regression.

Inventors:
MAO LI (CN)
MAO YU (CN)
XIAO YONGSONG (CN)
Application Number:
PCT/CN2016/082668
Publication Date:
November 23, 2017
Filing Date:
May 19, 2016
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
UNIV JIANGNAN (CN)
International Classes:
G06N3/00; G06N99/00
Foreign References:
CN106022465A2016-10-12
CN102708381A2012-10-03
Other References:
LUO, JUN ET AL.: "Artificial Bee Colony Algorithm with Chaotic-Search Strategy", CONTROLAND DECISION, vol. 25, no. 12, 31 December 2010 (2010-12-31), ISSN: 1001-0920
Attorney, Agent or Firm:
WUXI HUAYUAN PATENT AND TRADEMARK AGENCY (GENERAL PARTNERSHIP) (CN)
Download PDF: