A Genetic Programming-Driven Data Fitting Method

杨利霞

A Genetic Programming-Driven

( i8 Z) s4 C# B/ |

Data Fitting Method

) `2 M+ p6 a3 u: i% M

& Y6 [9 v/ X( a. e+ F# |  l; y
% A  o$ y. f! B) d" u& qData 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,
% f: `% S+ |8 F% i( [1 `. k; Nsuch as the polynomial model, we notice that a hybrid fifitting model with multiple types of function may have
a better fifitting result. Moreover, this also shows better interpretability. However, a perfect smooth hybrid
! B# Q* B- M6 I( n% ]8 _- Yfifitting 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
; B3 U& x: E% e4 i, z( j  S% \# F/ Emodel construction approach. In this approach, the model is expressed by an improved tree coding expression
3 T. T6 ]7 N- ~5 U4 X. rand constructed through an evolution search process driven by the genetic programming. In order to verify
the 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
( W2 L- H8 t$ n, @3 k" r. xaccuracy and interpretability.





  L# g6 E# j' @  {+ D