A Genetic Programming-Driven Data Fitting Method

杨利霞

A Genetic Programming-Driven

7 S) e' f5 {: g

Data Fitting Method

" T. Y6 Q; n  o; U3 V

' g# K8 m4 _% ]6 b6 B8 {, rData 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,
# v7 `7 h5 _1 Qsuch as the polynomial model, we notice that a hybrid fifitting model with multiple types of function may have
2 R- P0 M$ D, T) ba 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
4 {* k: Z4 x3 hmodel construction approach. In this approach, the model is expressed by an improved tree coding expression
8 X# y" W# j- m* E- dand constructed through an evolution search process driven by the genetic programming. In order to verify
. u8 d: \: w5 h9 e7 H, `9 j5 w1 Z; cthe 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.


9 b& J; W- W9 o  \5 `1 f# r
( |. s. J, P  i7 c8 C* B2 b
- o: P' C, W( X4 p' h7 g