A Genetic Programming-Driven Data Fitting Method

杨利霞

A Genetic Programming-Driven

Data Fitting Method

- ^5 i; ]: [7 J6 m


2 W0 o1 y& K6 Y! E7 CData fifitting is the process of constructing a curve, or a set of mathematical functions, that has
the best fifit to a series of data points. Different with constructing a fifitting model from same type of function,
8 \& T6 H! q# f8 P. a: J# nsuch as the polynomial model, we notice that a hybrid fifitting model with multiple types of function may have
" r: [$ y6 W* t" Q% q" A1 Qa better fifitting result. Moreover, this also shows better interpretability. However, a perfect smooth hybrid
1 H1 ?9 k4 T4 {5 J$ Efifitting model depends on a reasonable combination of multiple functions and a set of effective parameters.
That is a high-dimensional multi-objective optimization problem. This paper proposes a novel data fifitting
model construction approach. In this approach, the model is expressed by an improved tree coding expression
% {! J1 R! ]7 s+ X" R3 |" d1 |, oand constructed through an evolution search process driven by the genetic programming. In order to verify
8 H* ?; w  @% X3 T& r5 g& Q$ V. f) Lthe validity of generated hybrid fifitting model, 6 prediction problems are chosen for experiment studies. The
experimental results show that the proposed method is superior to 7 typical methods in terms of the prediction
6 Z" g: @& }/ q5 V& iaccuracy and interpretability


2 J9 D8 o0 i/ ~+ o. r$ f( Z

! T( c, S9 n; C0 V2 I