Research on chaos recognition of short time series with noise based on least square support vector machine - Details

author：

Xiaodong, Xiang (Xiaodong, Xiang.) ^[1] (Scholars：向小东) | Meiyan, Chen (Meiyan, Chen.) ^[2]

Indexed by：

EI Scopus

Abstract：

In　order　to　solve　the　chaos　recognition　problem　of　short　time　series　with　noise　in　reality,　a　new　method　of　data　expansion　combined　with　chaos　theory　and　least　square　support　vector　machine　is　proposed.　Then　the　fractal　dimension　and　the　largest　lyapunov　exponent　are　calculated　on　the　basis　of　expansion　data.　Finally　the　simulation　of　Henon　system　indicats　the　effectiveness　of　this　method,　and　explains　that　this　method　has　some　practical　value.　©　2011　IEEE.

Keyword：

Chaos theory e-government Fractal dimension Least squares approximations Lyapunov methods Support vector machines Time series

Community：

[ 1 ] [Xiaodong, Xiang]School of Management, Fuzhou University, Fuzhou, Fujian, China
[ 2 ] [Meiyan, Chen]School of Management, Fuzhou University, Fuzhou, Fujian, China

Reprint 's Address：

Email：

Xiangxiaodong2@yahoo.com.cn

Show more details

Version：