Is it possible to use recurrent neural network for classification type problems?

Hello,

i have NASA "CM1"/Software defect prediction dataset, this dataset includes 21 attributes and one defect class (True/False). I know, i can use feed-forward neural network on this dataset for defect prediction, But i want to use "Recurrent Neural Network" on this dataset, i also know Recurrent Neural Network are design for time series problem, However, is it possible i can also use "Recurrent Neural Network" for my "classification" problem? and this network will return good results?

Thanks

Answer 1

[Greg Heath]

It could happen; but don't count on it because there is no reason why it should be any better than without it.

Greg

Thank you for formally accepting my answer

Greg

[Ahmed iqbal]

We can say that? Recurrent Neural Network are extended version of feed-forward neural network, with TimeSeries support? you mean, we can still use Recurrent Neural Network for classification tasks, but we would not expect any significantly outperformed results.

[Greg Heath]

I don't know. However, I do know that if you could, there probably would be no advantage.