Font Size: a A A

Improved Flower Pollination Algorithm Extreme Learning Machine Classification Model

Posted on:2020-07-09Degree:MasterType:Thesis
Country:ChinaCandidate:C H LiFull Text:PDF
GTID:2428330623465357Subject:Software engineering
Abstract/Summary:PDF Full Text Request
Aiming at the problem of classification accuracy fluctuation caused by input layer weight and threshold random selection of multi-output extreme learning machine(MELM)classification model,a multi-classification model of extreme learning machine based on improved flower pollination algorithm(ACFPA)is proposed(CS-ACFPA-MELM).).First,for the flower pollination algorithm,the initial gamete position is optimized using the Tent mapping based on reverse learning.Introducing Tent chaotic search with small probability variation in global search.Add an adaptive operator to the local search.Convert the switching probability p into a function of the number of iterations.Secondly,construct a cost-sensitive fitness function to make the flower pollination algorithm better match the output of the MELM model.Finally,use the cost-sensitive fitness function to optimize the input weight and threshold of the extreme learning machine to improve the MELM.The classification performance of the model.In the experimental part,this paper uses a number of UCI data sets with different latitudes and different sample numbers to test the improved extreme learning machine classification model.It can be seen from the experiment that the accuracy of ACFPA algorithm is improved compared with FPA algorithm;improved flowers The pollination algorithm can effectively improve the classification accuracy of the extreme learning machine classification model.Compared with the FPA-MELM algorithm,the CS-ACFPA-MELM algorithm can improve the convergence speed of the model.At the same time,compared with the classic LS-SVM and CART classification algorithms,the CS-ACFPA-MELM model has obvious advantages in classification accuracy and Kappa coefficient as the number of samples increases.
Keywords/Search Tags:Extreme learning machine, flower pollination algorithm, chaotic search, cost sensitive, adaptive, reverse learning
PDF Full Text Request
Related items