标题: A Genetic Programming-Driven Data Fitting Method [打印本页] 作者: 杨利霞 时间: 2020-11-12 16:33 标题: A Genetic Programming-Driven Data Fitting Method
A Genetic Programming-Driven
$ {7 @2 u! r! |* ^3 m
Data Fitting Method
. K$ y8 @( c3 ?
2 x. ^( T8 M, x* l3 H6 M$ \ ; Q# z* N1 ~6 u3 Z' C H2 ^4 z& d " C( g; @+ n" |6 r6 ]& DData fifitting is the process of constructing a curve, or a set of mathematical functions, that has 5 ]4 V/ z* @ `the best fifit to a series of data points. Different with constructing a fifitting model from same type of function,' c6 K1 Y) J+ c# D2 H% {
such as the polynomial model, we notice that a hybrid fifitting model with multiple types of function may have; d: O# I, C0 e- Y, ^2 B1 }
a better fifitting result. Moreover, this also shows better interpretability. However, a perfect smooth hybrid% B( m/ b% _/ j! b4 _
fifitting model depends on a reasonable combination of multiple functions and a set of effective parameters.0 I+ w* i( }7 j4 h3 Z
That is a high-dimensional multi-objective optimization problem. This paper proposes a novel data fifitting 2 _3 r Q$ B: [. ?4 Smodel construction approach. In this approach, the model is expressed by an improved tree coding expression / R. M; U% G/ l, o7 [and constructed through an evolution search process driven by the genetic programming. In order to verify 0 U# } S, W. \+ v- [+ @$ c$ Gthe validity of generated hybrid fifitting model, 6 prediction problems are chosen for experiment studies. The5 @! b; I: h' O5 C" y
experimental results show that the proposed method is superior to 7 typical methods in terms of the prediction' n% |; {) e B/ ^) f/ M9 q
accuracy and interpretability9 l s& }) u/ J( X
: C ^) @2 j: k
2 c" v8 l: U" f
