Saleem,
Afroz and Biplob are thinking on taking up "emotional analysis of text" as the project work.
Hi guys,
As i said in the chat with Afroz. Text analysis is very complex and vast area of research. So it will be better if you guys can come up with synopsis of the specific problem you guys are trying to solve.
For the starters you guys need to get a general idea of natural language processing, text analysis, machine learning, model creation and training, parsing text.
So you see its a lot of work .. real hard to wrap in minor project.
Now lets talk about how to progress on this.
And than begin creating a model which will give better results for ur problem. And train it with data, huge data. you can easily find datasets on infochimps, freebase and others. Twitter tweets dataset shld be nice to analyse for emotional value.
For getting fast implementation, you can look at different opensource libraries for text analysis algorithms. python (NLTK), java(GATE).
Cheers,
On Fri, Jul 22, 2011 at 6:20 PM, afroz alam
<afroza...@gmail.com> wrote:
--
Kind Regards,
Afroz Alam
Computer Engineering