A Genetic Programming-Driven Data Fitting Method

杨利霞

A Genetic Programming-Driven

Data Fitting Method

" E; c8 i% c  Z2 l* I. w
- P: u; p8 B( Q2 g8 U% C  c

Data fifitting is the process of constructing a curve, or a set of mathematical functions, that has
& B0 G2 K% e  v: X5 `the best fifit to a series of data points. Different with constructing a fifitting model from same type of function,
* A$ c/ Y. B8 N) ~' Ssuch as the polynomial model, we notice that a hybrid fifitting model with multiple types of function may have
8 ^1 ^* L1 [/ ?0 e- \/ B, ^6 Ba better fifitting result. Moreover, this also shows better interpretability. However, a perfect smooth hybrid
9 \# @' q2 O3 C: M  n7 I+ }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
( h0 s7 q3 h3 [; f- f* Nand 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
0 J0 @: `" }1 M' D- xexperimental results show that the proposed method is superior to 7 typical methods in terms of the prediction
accuracy and interpretability

0 n5 x7 b4 Y+ O
" `& `+ \+ F( e5 Q  n( r# _
9 y7 y$ D% {7 h
0 V7 {4 k0 b. R" E3 W: ]- p
$ C, O% V5 ]# A! h: _8 ?