A Genetic Programming-Driven Data Fitting Method

杨利霞

A Genetic Programming-Driven

Data Fitting Method

: j$ b4 K; I) P- f  a% [2 I
* T+ X& L: `4 T1 x# i
Data 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,
such 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
fifitting 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
; M) y+ ^2 d& G: b( o/ aand constructed through an evolution search process driven by the genetic programming. In order to verify
5 k6 Q2 M7 b+ c5 X4 \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
) d0 k- c- b/ D: N3 d: Paccuracy and interpretability
5 v& E9 ^8 [; }4 E( w# c0 U. {
, p/ d. N, ~" U1 t+ s
8 [; |$ {5 A' I+ E4 R) H% e


+ }! J4 K# C3 x2 D, }5 S1 y1 {3 }/ e) H3 a