The neuralnet package in R is adequate for my requirements.

You may like to explore

Pylearn2 (http://deeplearning.net/software/pylearn2/#) : Top flexibility but difficult to approach in the beginning.

Theano-nets (https://github.com/lmjohns3/theano-nets): Very interesting and useful... maybe the first to try. Again it speaks Python.

HTH

In comparision to Weka, R is free and very easy to use s/w for machine learning.

SAS is commercialy very famous. Mathmatically researchers love to explore machne learning with Matlab.

The best one: Encog (http://www.heatonresearch.com/encog)

You may like having a look at http://www.kdnuggets.com/software/index.html

By default however I would explore any of the R packages as some of the commenter colleagues.

Try rapidminer

Use the AMORE package for R. Easy usage and highly configurable

scilearn python

JavaNNS of the University of Tübingen (version in Java of the SNNS)

http://www.ra.cs.uni-tuebingen.de/software/JavaNNS/welcome_e.html

Others:

https://decibel.ni.com/content/docs/DOC-17620

http://www.membrain-nn.de/main_en.htm

Good luck

There are many software that can be used:

Like Matlab, Synapse.. an other.

R is free and very easy to use s/w for machine learning.

Please check the link below for others:

http://www.kdnuggets.com/software/classification-neural.html

Good luck!

Thanks for everyone to provide insights about the topic.