A Genetic Programming-Driven Data Fitting Method

杨利霞

A Genetic Programming-Driven

Data Fitting Method

: i$ |8 x/ p" ]3 t! C9 e/ _3 D
Data fifitting is the process of constructing a curve, or a set of mathematical functions, that has
* B# W7 K1 s% mthe 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.
+ h" B! q6 X3 y( b% F4 @2 JThat is a high-dimensional multi-objective optimization problem. This paper proposes a novel data fifitting
! L* c" U7 w, ~  g6 y7 _' y: vmodel construction approach. In this approach, the model is expressed by an improved tree coding expression
8 J) c$ N: p+ Uand constructed through an evolution search process driven by the genetic programming. In order to verify
, n, v/ x: w# K7 d9 H  r) s1 W/ g. A) O! Qthe 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
  \. I6 l# d# Z( Q


$ I$ x- s" H, G% ]$ _3 j