How can I do a pattern recognition of a sequence of data using Neural Network?

I would like to compare sequences of data (a) with another one (b).

My objective is to see if each one (a) has partial or complete common sequence patterns with the second one (b).

To do so, I am trying to implement a Pattern Recognition Neural Network. I would like to divide each sequence (a) into smaller sequences and then introduce these small sequences into the Neural Network and obtain if they have a common pattern with (b).

However, when creating the Neural Network I do not know how to establish the Inputs and Targets. I would like to have the sequence (b) as an Input. The targets should be 1 in case of the sequence (2) and 0 when another sequence is shown. But I do not know how to implement it, I guess it is not going to be enough.

Does anyone have any idea of how could I do so?

Thank you very much in advance

[Greg Heath]

Before designing a classifier, I suggest that you cluster the data.

Greg

[MRE]

Excuse me Greg, I do not understand: why should I cluster the data? I mean, how is it going to help me to classify if my sequence (a) is into the sequence (b) or not?

Thank you for your reply

[Greg Heath]

In order to make comparisons you will have to fix a sequence length. Sequences of the same length can be clustered.

The problem with this is you have to have additional information to determine which lengths are reasonable to investigate.

Hope this helps.

Greg