A Genetic Programming-Driven Data Fitting Method

杨利霞

A Genetic Programming-Driven

* k* T& K3 r+ \$ m* Q6 b) }

Data Fitting Method

& s2 S# h* W# `

" T' V* P5 `1 I, K" sData 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
' ~- \. |% K6 O5 J; l( X- n) n7 l0 Qa better fifitting result. Moreover, this also shows better interpretability. However, a perfect smooth hybrid
$ i6 J/ N7 [5 |( |% H+ p! Qfifitting model depends on a reasonable combination of multiple functions and a set of effective parameters.
8 P0 T6 u" J! O7 vThat 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
9 j+ D* N/ k4 x, D, Nand constructed through an evolution search process driven by the genetic programming. In order to verify
" D* w0 l5 X9 d( h' ~' g6 Othe 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 @" g$ A& A' e0 G, w0 \4 t# O0 _; i