Content

Airfoil Self Noise

Code download link: Airfoil.zip


This example shows application of Kolmogorov-Arnold model to Airfoil Self-Noise Data Set. The data set is published in 2014. It has 1503 records with 5 features. The outputs (last column) are real numbers. Below is example of data format (first few lines): 
800;0;0.3048;71.3;0.00266337;126.201
1000;0;0.3048;71.3;0.00266337;125.201
1250;0;0.3048;71.3;0.00266337;125.951
1600;0;0.3048;71.3;0.00266337;127.591
2000;0;0.3048;71.3;0.00266337;127.461
The entire data set is divided into 3 clusters, 60% for training, 20% for selection and 20% for testing. The model is built 16 times, and tested on selection cluster, the model with best result then used on testing data. The accuracy metrics is Pearson correlation coefficient. Usual accuracy is near 95%. Below is example of program print out:



The result can be compared to Neural Designer benchmark, which can be found on their help page. They also report same accuracy near 95%.