A Genetic Programming-Driven Data Fitting Method

杨利霞

A Genetic Programming-Driven

  ~7 _* r2 m, y9 K

Data Fitting Method

* Z4 P8 |% S" I* M) l

Data fifitting is the process of constructing a curve, or a set of mathematical functions, that has
2 U% J! `* g) {/ |the best fifit to a series of data points. Different with constructing a fifitting model from same type of function,
' w. v4 z6 {0 c& x- K6 j7 R2 v0 Ksuch 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
' a6 f: u" h  Q. B2 [6 M0 ~' e* Vfifitting 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
1 c  \. X; _5 i. k9 {, p/ C; v- jmodel construction approach. In this approach, the model is expressed by an improved tree coding expression
! ~* d2 U! q% \4 mand 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
accuracy and interpretability
/ h  t% S# E) W/ Y8 T! y

  x1 T' d( t: h( e& ~

) u$ X% E: S& X7 }" F1 X
( [- C9 B9 S- M$ D# p: p7 ]2 r% V