A Genetic Programming-Driven Data Fitting Method

杨利霞

A Genetic Programming-Driven

Data Fitting Method

: B) W" m) X3 a1 T2 ^
+ B/ `; R  ?  q" u0 u$ v1 g

Data fifitting is the process of constructing a curve, or a set of mathematical functions, that has
: n0 `0 p$ d- i7 ~the best fifit to a series of data points. Different with constructing a fifitting model from same type of function,
! v( l& Q: b% T* _( tsuch 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
- c% z+ e) {; L/ H& y" gmodel construction approach. In this approach, the model is expressed by an improved tree coding expression
" D( E( R7 k" dand 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
: j$ p2 O2 `5 vexperimental results show that the proposed method is superior to 7 typical methods in terms of the prediction
8 g4 H' s. P9 y4 i4 i0 Taccuracy and interpretability