~ Nonlinear Adaptive Filtering as a Form of Artificial Intelligence ~

Mushroom Classification Example

Demo project, code and data code and data
This DEMO is simple comparison of discrete Urysohn model with Classification Tree. The data file is available for 30 years. It is mushroom classification. The matter of the data is list of quantized features for edible and poisonous mushrooms and the code must train AI to tell edible mushroom from poisonous by provided features for previously unseen mushroom. This is how the data look like:
p,x,s,n,t,p,f,c,n,k,e,e,s,s,w,w,p,w,o,p,k,s,u
e,x,s,y,t,a,f,c,b,k,e,c,s,s,w,w,p,w,o,p,n,n,g
e,b,s,w,t,l,f,c,b,n,e,c,s,s,w,w,p,w,o,p,n,n,m
p,x,y,w,t,p,f,c,n,n,e,e,s,s,w,w,p,w,o,p,k,s,u
e,x,s,g,f,n,f,w,b,k,t,e,s,s,w,w,p,w,o,e,n,a,g
For comparison I used Microsoft.ML library. It runs for about 60 sec (ten folds) and returned 57 errors out of 8124 records. 10 folds means the dataset was divided with ratio 0.9/0.1 into training and testing samples and testing set was rotated 10 times to cover all data. So the execution speed was about 6 sec per run. The Microsoft version can be found here:
https://www.c-sharpcorner.com/article/how-to-do-binary-classification-in-asp-net-core-using-ml-net/ My code with discrete Urysohn model runs for about two seconds, the logical part of the code is about 60 lines, the accuracy is 3 or 4 errors for unseen data out of 8124 for the same 10-fold validation concept.

Conceptually, it is same simple quantized Urysohn operator explained in introduction.