Pattern Recognition

Hello!

My question deals with Pattern Recognition problem.

I have a training data and a test data(so it is a supervised problem). The training data is a short pattern e.g. with 2 columns:

Col0: 1 0 0 1 1 1 0 0 0 0 1 0 1 0 1 1 1 0 0 0 1 0 1 1 0 0 0 0

Col2: “Warning” “Warning”… in each case, with as many rows as Col0 with the string

The test data would then be:

Col0: 100 101 102 103…… As many rows as Col1 containing these Time values

Col1: from Row0 to Row100 “random()”

          The following rows: 1 0 0 1 1 1 0 0 0 0 1 0 1 0 1 1 1 0 0 0 1 0 1 1 0 0 0 0

          The following rows: “random()”

Here is my workflow: ( and attached as well )

 

Fuzzy rules can be used as well. The Learner has got as Target Col1 of the training data.

So what I expect is that the Predictor gives as an output “Warning” only for the rows where it recognizes the training pattern.

RESULT that I get here: “Warning” signal in every single row of the “Winner” column.

 

Possible remedy: I enter some additional random values in the training data so as to have:

Col0: 1 0 0 1 1 1 0 0 0 0 1 0 1 0 1 1 1 0 0 0 1 0 1 1 0 0 0 0 0.52 9 456 0.6 3

Col2: “Warning” “Warning”… ………………………”Warning”   0     0   0     0  0

RESULT: some more “0”s but it’s not perfect, and it depends on the values I put additionally.

 

Is there some other way of solving the problem? Because you can guess that with many inputs either in the training or in the test data, the problems still remain.

Or would there be a completely different type of workflow that is much better?

Thanks in advance!

Heej.